Deconstructing History

Deconstructive readings of history and sources have changed the entirediscipline of history. And in this second edition of Deconstructing His-tory, Alun Munslow examines history in what he argues is a post-modern age. He provides an introduction to the debates and issues ofpostmodernist history. He also surveys the latest research into the rela-tionship between the past, history and historical practice, as well asforwarding his own challenging theories.

In this fully up-to-date second edition, Munslow:

• discusses the limits of conventional historical thinking and practice

• assesses afresh the claims of history as a form of ‘truthfulexplanation’

• examines the arrival of ‘experimental history’ and addresses itsimplications for a radical rethinking of the discipline.

Including a fully updated glossary and bibliography, Munslow maps thephilosophical field, outlines the controversies involved and assessesthe merits of the by now familiar deconstructionist position.

Alun Munslow is Visiting Professor of History and Historical Theory atthe University of Chichester and UK editor of Rethinking History: TheJournal of Theory and Practice. He is co-editor of Experiments inRethinking History (2004) with Robert A. Rosenstone and of TheNature of History Reader (2004) with Keith Jenkins and is author ofThe Routledge Companion to Historical Studies (2nd edition, 2006), allpublished by Routledge. He has also published The New History (2003).

Deconstructing HistorySecond Edition

Alun Munslow

First published 1997by Routledge2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

Simultaneously published in the USA and Canadaby Routledge270 Madison Ave, New York, NY 10016

Routledge is an imprint of the Taylor & Francis Group, an informa business

© 1997, 2006 Alun Munslow

All rights reserved. No part of this book may be reprinted orreproduced or utilised in any form or by any electronic,mechanical, or other means, now known or hereafterinvented, including photocopying and recording, or in anyinformation storage or retrieval system, without permission inwriting from the publishers.

British Library Cataloguing in Publication DataA catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication DataMunslow, Alun, 1947– .

Deconstructing history / Alun Munslow.—2nd ed.p. cm.

Includes bibliographical references and index.ISBN 0–415–39143–1 (hardback : alk. paper) — ISBN 0–415–39143–X

(pbk. : alk. paper)1. History—Methodology. 2. Historiography. 3. History—Philosophy.4. Narration (Rhetoric) 5. Literature and history. I. Title.D16.M963 2006907.2–dc22 2005028479

ISBN10: 0–415–39143–1 (hbk)ISBN10: 0–415–39144–X (pbk)ISBN10: 0–203–96990–1 (ebk)

ISBN13: 978–0–415–39143–6 (hbk)ISBN13: 978–0–415–39144–3 (pbk)ISBN13: 978–0–203–96990–8 (ebk)

This edition published in the Taylor & Francis e-Library, 2006.

“To purchase your own copy of this or any of Taylor & Francis or Routledge’scollection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.”

Contents

Acknowledgements vi

1 Introduction 1

2 The past in a changing present 19

3 History as reconstruction/construction 39

4 History as deconstruction 61

5 What is wrong with deconstructionist history? 82

6 What is wrong with reconstructionist/constructionist history? 107

7 Michel Foucault and history 129

8 Hayden White and deconstructionist history 149

9 Conclusion 175

Glossary 192

Guide to further reading 207

Notes 216

Index 240

Acknowledgements

This book is the result of a prolonged period of teaching and thinkingabout the ways in which it is possible to write the past. As a con-sequence there are many colleagues who, often unwittingly perhaps,have made me reassess my thinking as a historian. I am grateful to themall. As ever, my final thanks are to Jane, who has always known thathistory is a story.

1 Introduction

APPROACHING HISTORY

It is my intention to navigate through the central debate to be found inhistory today, viz. the extent to which history, as a discipline, canaccurately recover and represent the content of the past, through theform of the narrative. Put plainly, to what extent is the narrative orliterary structure of the history text an adequate vehicle for historicalexplanation, and what implications can we draw from our answer? It isnow commonplace for historians, philosophers of history and othersinterested in narrative to claim we live in a postmodern age wherein theold modernist certainties of historical truth and methodological objec-tivity, as applied by disinterested historians, are challenged principles.Few historians today would argue that we write the truth about thepast. It is generally recognised that written history is contemporary orpresent orientated to the extent that we historians not only occupy aplatform in the here-and-now, but also hold positions on how we seethe relationship between the past and its traces, and the manner inwhich we extract meaning from them. There are many reasons, then,for believing we live in a new intellectual epoch – a so-called post-modern age – and why we must rethink the nature of the historicalenterprise to meet the needs of our changed intellectual beliefs andcircumstances. Later in this chapter I will pose some basic questionsabout the nature of history, not least the fundamentally changednature of how we come to understand the past as a body of knowledgefrom which we can derive a meaning for it. As we shall see, it is pre-cisely this situation of how we constitute knowledge about the past thatdirectly a!ects the nature of the meaning we impose upon it. Historycan no longer legitimately be viewed as simply or merely a matter ofthe discovery of the story of the past, the detection of which will tellus what it means. This belief results from a debate on the nature of

knowing that began well over one hundred years ago in the nineteenthcentury.

What are these changed circumstances that justify the claim that welive in a postmodern age? First, the claim is not being made that post-modernism is a particularly new perspective or position arrayed againstother old positions or perspectives about how we gain knowledge of thereal past (or present). Postmodernism is, rather, the changed and con-temporary condition under which we gain knowledge. Among the keyprinciples of this new condition of knowing are the broad doubts thatnow exist about the accurate representation of reality. Indeed, post-modernism is not particularly new if we think about the self-reflexivityof the period supposed to exist prior to it.

Indeed, the term postmodernism is actually somewhat misleading.You will note I use the term un-hyphenated in this book. Rather than‘post-modernism’ which is often the way it is described, I prefer to thinkof our present intellectual age not as something that came after (hencepost-) but which is rather a transmutation of modernism. ‘Post-modernism’ has often been deployed to mean the arrival of a new set ofconditions for knowing when it seems more appropriate to say modern-ism has now become fully aware of its own in-built critique of knowing.So, as we shall see, much that we refer to as postmodern (un-hyphenated) is in fact modernism’s re-evaluation – especially in the lastthirty years or so – of its own principles.

One of the main points about the Age of Enlightenment modernismfrom the seventeenth and eighteenth centuries and through the nine-teenth and twentieth centuries was its self-consciousness in asking ques-tions about how we know what we know. In a peculiar sense, perhapsmodernism was always going to end up fundamentally critiquing itself.Maybe postmodernism was the inevitable consequence of modernism?We will see how this a!ects the study of the past throughout the rest ofthis book, but it is important from the start to recognise that historywas always going to be in the forefront of this modernist will to self-criticism. It is as a result of this postmodern condition for knowingthat history, as a discipline, has always been particularly susceptible todebates about its nature.

This book is called Deconstructing History because at its core is mybelief that history must be reassessed at its most basic level. It is notenough merely to criticise historical method, but rather to ask can pro-fessional historians be relied upon to reconstruct and explain the pastobjectively by inferring the ‘facts’ from the evidence, and who, afterall the hard work of research, will then write up their conclusionsunproblematically for everyone to read?

2 Introduction

Even if, as many might argue, history has never been nor is nowprecisely as positivist a research process, or as unreflective a literaryundertaking as that description suggests, the crude empiricist or recon-structionist emphasis on the historian as the impartial observer whoconveys the ‘facts’ is a paradigm (defined as a set of beliefs about howto gain knowledge) that obscures history’s real character as a literaryundertaking. I will argue that the genuine nature of history can beunderstood only when it is viewed not solely and simply as an objectiv-ised empiricist enterprise, but as the creation and eventual impositionby historians of a particular narrative form on the past: a process thatdirectly a!ects the whole project, not merely the writing up stage. Thisunderstanding, for convenience, I shall call the deconstructive con-sciousness. This use of the term is not to be confused with its originaluse by French cultural theorist Jacques Derrida, who employed theterm more narrowly to mean the process whereby we grasp the meaningof texts without reference to some originating external reality. Thedeconstructive consciousness not only defines history as what it palp-ably is, a written narrative (the textual product of historians), but ad-ditionally, and more radically, suggests that narrative as the form ofstory-telling may also provide the textual model for the past itself. Rec-ognising the literary dimension to history as a discipline does not meanthat we cannot ask ourselves is it only our lived experience that is retoldby historians as a narrative, or as historical agents do we experiencenarratives – as people in the past? In other words, does the evidencereveal past lives to be story-shaped, and can we historians retell thenarrative as it actually happened, or do we always impose our ownstories on the evidence of the past?

Whatever we decide, it follows that history cannot exist for the readeruntil the historian writes it in its obligatory form: narrative. What do Imean by narrative? When we explain in history we place its contents asevents in a sequential order, a process usually described as the telling ofa story. No matter how extensive are the analytical apparatuses bor-rowed from the social sciences and brought to bear on the past, his-tory’s power to explain resides in its fundamental narrative form. Asthe pro-narrative philosopher of history Louis Mink said in the early1960s: ‘Where scientists . . . note each other’s results, historians . . . readeach other’s books.’1 So far as this book is concerned, the reality of thepast is the written report, rather than the past as it actually was. I willargue that history is the study not of change over time per se, but thestudy of the information produced by historians as they go about thistask. In Deconstructing History I am attempting to highlight the essen-tially literary nature of historical knowledge and the significance of its

Introduction 3

narrative form in the constitution of such knowledge. In our con-temporary or postmodern world, history conceived of as an empiricalresearch method based upon the belief in some reasonably accuratecorrespondence between the past, its interpretation and its narrativerepresentation is no longer a tenable conception of the task of thehistorian. Instead of beginning with the past we should start with itsrepresentation, because it is only by doing this that we challenge thebelief that there is a discoverable and accurately representable truthful-ness in the reality of the past.

SOME BASIC QUESTIONS ABOUT THE NATURE OF HISTORY

Four specific questions about the nature of history flow from the beliefthat history as it is lived and written is structured as much by its form asby its content. Although we can distinguish these questions for thepurpose of listing, in practice it is very di"cult to keep them separate.

• Can empiricism legitimately constitute history as a separate epistem-ology?

• What is the character of historical evidence and what function does itperform?

• What is the role of the historian, his/her use of social theory, andthe construction of explanatory frameworks in historical under-standing?

• How significant to historical explanation is its narrative form?

These questions prompted the writing of this book and lie at the heartof the status crisis besetting history today.

Epistemology

The first question confronts the basic issue about history as a form ofknowledge: is there something special in the methods deployed by his-torians to study the past that produces a reliable and objective knowl-edge peculiar to itself, and which makes it possible to argue that there issuch a thing as a discipline of history at all? Historical knowledge, as itis usually described, is derived through a method – called a practice bythose who believe in the possibility of an accurate understanding of thepast – that flows from its techniques in dealing with the traces ofthe past. The most basic function of the historian is to understand, andexplain in a written form, the connections between events and humanintention or agency in the past. Put another way, the historian has towork out some kind of method or means whereby he/she can grasp the

4 Introduction

relationship between knowledge and explanation in order to find thefoundation of truth, if one exists.

One method would be to imitate the natural sciences, and althoughthere has always been a large minority following among historians(especially among those with a positivist or social science training) forthis flattery, it has never achieved a dominant methodological status.History cannot claim to be straightforwardly scientific in the sense thatwe understand the physical sciences to be because it does not share theprotocol of hypothesis-testing, does not employ deductive reasoning,and neither is it an experimental and objective process producingincontrovertible facts. Moreover, the better we do it does not guaranteewe will get closer to the truth. Scientific method works on the assump-tion that data are connected by a universal explanation, and con-sequently the scientist selects his/her data according to this belief. Thehistorian, however, selects his/her data because of his/her interest in aunique event or individual acting intentionally in response to circum-stances. Evidence is chosen for what it can tell us about that uniqueevent or individual, rather than any and every event within a generalcategory being explained.

What particular consequences flow from this for history as an epis-temology, or special form of knowledge?2 Can we gain genuine and‘truthful’ historical descriptions by simply following the historian’s lit-erary narrative – her or his story? This is certainly the opinion of severalcommentators. The British theorist of history M.C. Lemon considersthat the ‘very logic’ of history as a discipline revolves ‘around therationale of the narrative structure’.3 In respect of what peculiarly con-stitutes historical explanation, Lemon argues that its essence lies in themanner in which historians account ‘for occurrences in terms of thereasons individuals have for their conduct’. In other words, history canbe legitimately defined as the narrative interpretation and explanationof human agency and intention.4 The special character of narrative thatmakes it so useful to historians is, as Lemon points out, its ‘this hap-pened, then that’ structure which also, of course, is the essence of his-torical change. It is a process that saturates our lived experience. Inother words, the past existed and will exist as knowledge transmitted tous according to the basic principles of narrative form.

What, then, is the relationship of history to its closest neighbour,literature? The bottom line seems to be one of referentiality. I take thisto mean the accuracy and veracity with which the narrative relates whatactually happened in the past. As Lemon argues, while literature is notwholly ‘devoid of referentiality’, it is ‘not referential in the same man-ner’ as the historical text.5 It follows that, like literature, the past and

Introduction 5

written history are not the same thing.6 Not recognising this permits usto forget the di"culties involved in recreating the past – something thatdoes not exist apart from a few traces and the historians’ narrative.Because we cannot directly encounter the past, whether as a politicalmovement, economic process or an event, we employ a narrative fulfil-ling a two-fold function, as both a surrogate for the past and as amedium of exchange in our active engagement with it. History is thusa class of literature.

The most basic assumption that informs my book is that the past isnegotiated only when historians represent it in its narrative form andthat historical interpretation should not close down the meanings of thepast to pursue what at best must remain an ersatz ‘truth’. Indeed, weought to be more open to the possible meaninglessness or sublime char-acter of the past. Although mainstream empiricists may dispute it, Ishall argue that there cannot be any unmediated correspondencebetween language and the world as a discoverable reality. Of course,even if this is the case, it does not stop us from asking, although wecannot provide a definitive answer, is it possible that the past unfoldedas a particular kind of narrative the first time around and can werecover it more or less intact, or are we only selecting and imposing anemplotment or story line on it derived from our own present? Are stor-ies lived in the past or just told in the present? Do we explain our lives atthe time like the unfolding of a story? The most important question,then, is not the dog-eared modernist one of whether history is an accur-ate science, but the postmodernist one of how and why when we writeabout the past, we cast it in a particular narrative form. Further, howadequate is the cognitive power of narrative? What is its capacity toexplain the past plausibly?

Just as it is impossible to have a narrative without a narrator, wecannot have a history without a historian. What is the role of the his-torian in recreating the past? Every history contains ideas or theoriesabout the nature of change and continuity as held by historians – someare overt, others deeply buried, and some just poorly formulated. Thetheories of history mustered by historians both a!ect and e!ect ourunderstanding about the past, whether they are explicit or not. To theextent that history is a narrative interpretation built in part out of thesocial theories or ideological positions that historians invent to explainthe past, history may be defined essentially as a language-based manu-facturing process in which the written historical interpretation isassembled or produced by historians. As the pro-narrative philosopherof history Arthur Danto put it, ‘to tell what happened . . . and toexplain why . . . is to do one and the same thing’,7 or in the words of

6 Introduction

Lemon, the historian regularly encounters questions of ‘selection, rel-evance, significance and objectivity’ in his/her description of events.8 Iwill suggest, therefore, that history is best viewed epistemologically as aform of literature producing knowledge as much by its aesthetic ornarrative structure as by any other criteria. In addition, as we acknowl-edge history’s literary and fabricated character, I shall also address thepast as a narrative, as well as describe it in narrative.

Evidence

The second question concerns the raw materials in the history indus-try’s manufacturing process – the traces or evidence of the past. Weshould be beginning by now to see that because of the central role oflanguage in the constitution of knowledge, or historical understandingis as much the product of how we write as well as what we write, so thathistory’s so-called raw ‘facts’ are likewise presented either wholly or inlarge part to us in a written or literary form. Even raw statistics have tobe interpreted in narrative. If you, as a student of history, were asked togive an example of a historical ‘fact’, the normal response is to quote anincontrovertible event or description that everyone agrees upon. Thatslavery was the ultimate cause of the American Civil War is clearly notsuch a ‘fact’. It is a complex interpretation based on the relating ofdisparate occurrences, statistical data, events and human intentionstranslated as actions involving outcomes. But if we say in cold factualterms that the American President James Madison was ‘small of stature(5 feet, 4 inches; 1.62 metres), light of weight (about 100 pounds; 45kilograms), bald of head, and weak of voice’ this seems unproblematic– Madison was or wasn’t this tall, was or wasn’t slight, was or wasn’tbald, was or wasn’t weak voiced. The important point, however, is themeaning that these ‘facts’ about Madison produce in the mind of thereader, rather than the inherent veracity of the ‘facts’ themselves.

Because he was short, slight, bald and had a squeaky voice, does thisincline us towards an interpretation that he was weak, could nottherefore hold his cabinet together, and eventually became a dupe ofNapoleon?9 History is about the process of translating evidence intofacts. You and I as historians do this. Even when straight from the dustyarchive, the evidence always pre-exists within narrative structures and isfreighted with cultural meanings – who put the archive together, why,and what did they include or exclude? ‘Facts’ are literally meaningless intheir unprocessed state of simple evidential statement. The evidence isturned into ‘facts’ through the narrative interpretations of historians;but facts usually already possess their own narrators, and then they gain

Introduction 7

further meaning when they are organised by the historian as strands ina story producing a particular, appealing, followable, but above all aconvincing relationship. Historical interpretation is the written expla-nation of that perceived relationship.

As we can see, ‘facts’ are never innocent because only when used bythe historian is factual evidence invested with meaning as it is correlatedand placed within a context, sometimes called the process of colliga-tion, collation, configuration or emplotment, which then leads thehistorian to generate the ‘facts’.10 Traditionally, this process of contex-tualisation is undertaken by the historian as part of the process ofinterpretation as he/she relates masses of apparently unconnected datawith an eye to producing a meaning. The evidence of the past is pro-cessed through inference, with the historian construing a meaning byemploying categories of analysis supposedly determined by the natureof the evidence. The traces of the past are thus traditionally viewed asempirical objects from which to mine the meaning, or as sources out ofwhich social theories of explanation can be constructed.

However, this positioning or organising of the evidence in relation toother examples – what I choose to call the process of emplotment – iswhere the historian’s own views and cultural situation usually emerge.In writing history it is impossible to divorce the historian from theconstitution of meaning through the creation of a context, even thoughthis is seemingly and innocently derived from the facts. It is at this pointthat the historian unavoidably imposes him/herself on the past, whetherit be through the apparently wholesome practice of mining the evidencefor its true meaning, or more obviously through the creation and use ofsocial theories, but most importantly, I would suggest, because of theemplotment or story-line (narrative structure) deployed to facilitateexplanation and historical interpretation. I will examine the implica-tions of the role played in writing history by the evidence and ourrepresentation of it. Evidence is there for us to infer meaning from andthus create historical knowledge. However, the inference of meaningemerges as we organise, configure and emplot the data. It does not, Iwould argue, just turn up or suggest itself as the only or most likelyconclusion to draw.

Theories of history: constructing the past

The third question in this debate comes out of the belief, held by hard-core empiricists, that history is a practice founded on the objectivereconstruction of the facts, through which we get close to what actuallyhappened in the past. This is what the English philosopher of history

8 Introduction

R.G. Collingwood called ‘naïve realism’, and it is based on the idea thatexperience can be the object of historical knowledge.11 In order to sus-tain this position, such empiricists deny that historians should interveneor impose on the past by suggesting that they must be not onlyimpartial and objective in their treatment of the evidence, but also thatthey should reject social theory models in interpreting the past. Thislatter process is viewed by them as a crude construction or invention ofthe past.

However, since the 1920s, social and cultural history has been popu-lar because it demands the construction of explanations of how post-industrial society has been able/unable to cope with the massive socialchanges that have occurred in the train of capitalist industrialisation.This modernisation process could not be explained without recourse toa new and utilitarian kind of history in which historians played anactive role in its construction. They play this role either by empathicallyrethinking the thoughts of people in the past to ascertain their inten-tions, or by constructing social theory explanations rather than justwaiting for them to suggest themselves. Hard-core empiricists (Colling-wood’s naïve realists) today embrace the idea that historians must notyield to this twin siren call to justify historical interpretations byimagining or empathising with past historical actors, nor less constructuniversal explanatory theories (usually described today as meta-narratives) that can explain the past. Such empiricists refuse to acceptthe changing character of contemporary thought, not least what hasnow become a commonplace argument among the majority of his-torians, that historical knowledge is not objective but has upon it thefingerprints of its interpreters.

As twentieth-century Western society has experienced total war,social, political and ecological revolutions, and new technology, thegrowing need has been to make the past intelligible to the present, andthat means historians speculating on the causes of change, the nature ofcontinuity, and the endless possibilities in the past. Such speculationscannot simply rely upon empathy or its corollary historicism – seeingthe past in its own context or terms. Although the most obviousexample of twentieth-century constructionism is the Marxist school ofhistory that stresses the social theory of class exploitation as the modelof historical change, the advent in France in the 1920s of the Annalesschool of historiography also produced a parallel constructionist-socialscience-inspired history that proposed alternative behavioural anddemographic theories. From the 1970s an anthropologically indebtedsocial history has also emerged that challenges class as the major con-struct in historical explanation in favour of taking single events and

Introduction 9

decoding them for their wider cultural significance. In addition, themodernisation school has stressed the benefits of model-making tocomparative history. The New Economic History of the 1960s and1970s emphasised quantification. Constructionism has thus been sub-ject to fashions or trends.

Sociological and anthropological constructionism is one of the keysources of what has become known as the New Cultural History andwhat I shall designate as deconstructionist history. As a variant of con-structionism, the New Cultural History works on principles derived notonly from anthropology but also from the broader intellectual move-ment of post-structuralism which itself emerged from literary criticaltheory in the 1970s. Deconstructionist history regards the past as acomplex narrative discourse, but one, as the French cultural critic andhistorian Michel Foucault has pointed out, that accepts that representa-tion is not a transparent mode of communication that can adequatelycarry understanding or generate truthful meaning. Deconstructionisthistory is a part of the larger challenge to the modernist empiricistnotion that understanding emanates from the independent knowledge-centred individual subject designated variously as Man, humanity, theauthor or the evidence. As already noted, our postmodern condition forknowing has meant that the discipline of history debates its own natureas much as it does the meaning of the past. The most recent develop-ment has been the emergence of a ‘new empiricism’ that has ac-knowledged the postmodern critique, especially of history’s discursiveconstruction.12 Part of this recognition has been to emphasise thatempiricism has never been naïvely accepted. But as the term ‘new’ sug-gests, there is also an acknowledgement of the discursive or linguisticturn that indicates a degree of disillusionment with a realist view oflanguage and representation. Nevertheless, there remains a desire toretain empiricism, though be it in some kind of modified form, as thebedrock of history. In other words, there is, among new empiricists, adesire to argue that the correspondence theory of knowledge stillworks, though it is now opened up to the potential meanings in the past.No new empiricist is anti-realist, although the historian Carla Hessedescribed new empiricism in those terms.13 They are actually realistswho see empiricism as not containing a necessary or given meaning.This is the key notion behind the New Cultural History.

New Cultural Historians are increasingly moving towards this newempiricism. They are not epistemologically sceptical but they are epis-temologically self-conscious. This manifests itself in several di!erentreactions to modernist history. New Cultural Historians – dependentupon the personal proclivities of the individual – tend to be

10 Introduction

anti-representationalist. They are happy to accept teleological expla-nations if they are held to further certain ethical considerations (such asthe recovery of gender or race).14 This constitutes the emergence of theso-called ethical turn that has been increasingly significant in the pastdecade or so. There is a willingness to confront notions of time, whichfor modernists is resolutely linear. They accept and can work within theconcept of the fabrication rather than the discovery of meaning. Suchhistorians are often willing to work with the idea of history as a truth-e!ecting rather than a truth-acquiring discipline. They acknowledge thenarrowness of the boundaries between fact and fiction. They can be‘experimental’. They will explore the troubled relationship betweenform and content. They will be willing to work with a discipline that ishistoricist as well as linguistically constructed recognising the past isinextricably bound up with the present and its appropriation has neverbeen naïvely empiricist.

Modernist empiricism is in crisis because of the objection that mean-ing is generated by socially encoded and constructed discursive prac-tices that mediate reality so much so that they e!ectively close o! directaccess to it. This situation is compounded when language is considerednot to be a pure medium of representation. Is it any longer possible towrite history when not only are we looking at it through our con-structed categories of analysis – race, class, gender – but the narrativemedium of exchange itself confounds the realist and empiricist depen-dence upon what one commentator has called an ‘adequate level of cor-respondence between representations of the past and the past itself ’ asit once actually existed?15

The leading practitioner of the narrative or rhetorical version ofconstructionism remains the American philosopher of history HaydenWhite. White insists that history fails if its intention is the modernistone of the objective reconstruction of the past simply according to theevidence. It fails because the process involved is the literary one ofinterpretative narrative, rather than objective empiricism and/or socialtheorising. This means that writing history requires the emplotment ofthe past not just as a way of organising the evidence, but also takinginto account the rhetorical, metaphorical and ideological strategies ofexplanation employed by historians. The study of rhetoric as the modeof historical explanation is summarised in the claim that history isliterary artifact, as White says, as much invented as found.16

Introduction 11

History as narrative

Because history is written by historians, it is best understood as a cul-tural product existing within society, and as a part of the historicalprocess, rather than an objective methodology and commentary outsideof society. This brings us to the fourth key question – posed by Whitealong with Collingwood, and more recently by Louis Mink and ArthurDanto – what is the significance of narrative in generating historicalknowledge, and what is its relationship to the previous three questions?But, first of all, what do we mean when we talk of historical narrative?The modernist empiricist historical method handed down from thenineteenth century requires and assumes historical explanation willemerge in a naturalistic fashion from the archival raw data, its meaningo!ered as interpretation in the form of a story related explicitly,impersonally, transparently, and without resort to any of the devicesused by writers of literary narratives, viz., imagery or figurative lan-guage. Style is deliberately expunged as an issue, or relegated to a minorproblem of presentation. This vision of the history as a practice failsto acknowledge the di"culties in reading the pre-existing narrativeconstituted as evidence, or the problems of writing up the past.

We historians employ narrative as the vehicle for our reports butusually neglect to study it as an important part of what we do. For mostanalytical philosophers of history the essence of historical understand-ing is the ability to recognise, construct and follow a narrative, that is astory based upon the available evidence. A historical narrative is a dis-course that places disparate events in an understandable order: asLemon says, ‘this happened then that’. Such a narrative is an intellig-ible sequence of individual statements about past events and/or theexperiences of people or their actions, capable of being followed by areader while he/she is pulled through time by the author towards theconclusion. All such narratives make over events and explain why theyhappened, but are overlaid by the assumptions held by the historianabout the forces influencing the nature of causality. These might wellinclude individual or combined elements like race, gender, class, culture,weather, coincidence, geography, region, blundering politicians, and soon and so forth. So, while individual statements may be true/false,narrative as a collection of them is more than their sum. The narrativebecomes a complex interpretative exercise that is neither conclusivelytrue nor false.

The commonsense version of the general empiricist and reconstruc-tionist position on the essential role of narrative is well described by thephilosopher W.B. Gallie:

12 Introduction

Historical understanding is the exercise of the capacity to follow astory, where the story is known to be based on evidence and is putforward as a sincere e!ort to get at the story . . .17

Gallie is suggesting that the actual events as they really occurred inthe story of the past have a striking resemblance to the shape of thenarrative eventually produced by the historian – the narrative is found(discovered?) by the historian in the events themselves and then repro-duced. The narrative here has referentiality. While philosophers of his-tory like Keith Jenkins, Louis Mink and Hayden White believe that wedo not live stories but only recount our lived experience in the storyform, the American philosopher of history David Carr supports Gallieand French philosopher of history Paul Ricoeur in holding that there isa basic continuity or correspondence between history as it is lived (thepast) and history as it is written (narrated).18 Are we justified in claim-ing that because our lives are narrativised and written history is a text,then surely the past itself conforms to the structure of narrative? Whitereverses the argument – the narrative does not pre-exist but a narrativeis invented and provided by the historian. Consequently, there are manydi!erent stories to be told about the same events, the same past. Whilestill constrained by what actually happened (historians do not inventevents, people or processes), as the French historian Paul Veyne sug-gests, the meaning of history as a story comes from a plot, which isimposed, or, as Hayden White insists, invented as much as found by thehistorian.19

The argument runs that just as there are no grounds for believing thatan empiricist methodology can guarantee an understanding of the pastas it actually was, neither is there a discoverable original emplotment.However, the self-reflexive and self-conscious historian may argue thatit is possible to o!er an interpretation that, although not claiming to bethe true narrative, is nevertheless a plausible and therefore quite accept-able rendering of it. The range of emplotments upon which people inthe past, and the historian, draw, though wide because of the combin-ations possible, is formally limited to the four main kinds – romance,tragedy, satire and comedy. This is no di!erent to other narrators whooperate in the realm of fiction. Of equal significance to the narrativeemplotment, however, is the dimension of figurative description orstyle.

Historical story-telling, like all other kinds, employs the four primaryfigurative devices known as tropes. These are more commonly known asthe four primary figures of speech: metaphor, metonymy, synecdocheand irony, and their use constitutes what is called the troping process.

Introduction 13

Troping means turning or steering the description of an object, event orperson away from one meaning, so as to wring out further di!erent, andpossibly even multiple, meanings. When we use these four master tropeswe describe objects, events, persons or intentions, in terms of otherobjects, events, persons or intentions, according to their similarities ordi!erences, or characterise them by substituting their component partsfor the whole – like hands representing workers, the one element oraspect representing the essence of the other (as a synecdoche), or sailsfor ships, the one aspect again representing the other, but now in a part–whole relationship (to be read as a metonymy). Metaphor is, however,the most basic kind of trope, with metonymy, synecdoche and irony assecondary kinds. Metaphor refers to one thing by denoting another soas to suggest that they share a common feature. To deny literal meaningis to use irony. Troping is as crucial to the writing of history as it is toother forms of literature because it permits us to create meanings thatdi!er from those of colleagues and to disrupt the expectations ofreaders di!erent to those anticipated.

Later I shall examine emplotment and troping in White’s formalmodel and his argument that there is no continuity between livedexperience and its narrative representation, that narrative as a form ofhistorical explanation is ultimately inadequate, and that writing historyis also an unavoidably ideological act. Narrative is normally deployed,therefore, not to defend the correspondence theory of empiricism somuch as to act as its vehicle – getting at the story – but always at theexpense of the historical sublime. By this, White means the celebrationof the undiscoverable, possibly meaningless, and open-ended nature ofthe past. Such a meaninglessness is the only invitation that potentiallyoppositional and dissenting groups of historians may get to challengecertaintist (e.g. fascist) history. They, and we, empower ourselves whenwe can find no objective certainty in the past – in the sense of a factualcorrespondence of evidence with Truth – that can be used to validatethe authority of those in power over us.20 From a strictly philosophicalpoint of view, the existence of a past reality does not in itself verify thecorrespondence theory, since it does not mean the truth of past eventscan be found in any correspondence between the word and the world asstatements of past reality. Paradoxically, most historians, even leftistdissenting ones, prefer to believe it does.

Michel Foucault challenges this by arguing that the idea of Man(Man = historian for our purposes) is not able to stand outside societyand history and thus generate objective and truthful knowledge. Heconcludes (as does White) that language is an ideologically contamin-ated medium, and what it can and cannot do is dependent upon the use

14 Introduction

to which it is put, and for what social and political purposes – usually tomaintain or challenge systems of authority and views of what is right orwrong, allowed or banned. As he says, ‘ “Truth” is to be understood asa system of ordered procedures for the production, regulation, distribu-tion, circulation, and operation of statements. “Truth” is linked [to the]. . . statements of power, which produce and sustain it.’21 Foucault ispointing out here how historical agents – you and me – become con-federates in our own subjectivity rather than just victims. Through thefunctioning of language we cannot avoid being placed in subject posi-tions where the repression of the word fixes us all – like moths pinned toa collector’s board. In this sense, the organised study of the past (his-tory as a discipline which includes both its meanings: as a profession,and as a collection of methodological practices) as an organised narra-tive is itself founded on the dispensation of authority/power in con-temporary society. How we write history is as open to the uses andmisuses of power as any other narrative.

Written history is always more than merely innocent story-telling,precisely because it is the primary vehicle for the distribution and use ofpower. The very act of organising historical data into a narrative notonly constitutes an illusion of ‘truthful’ reality, but in lending a spuri-ous tidiness to the past can ultimately serve as a mechanism for theexercise of power in contemporary society. As White suggests, evenwhen we acknowledge and describe the messiness of the past, the veryact of narration imposes an unavoidable ‘continuity, wholeness, closureand individuality that every “civilised” society wishes to see itself asincarnating’.22 All historical narrative is thus subject to the complex andsubtle demands of ideology, and in its turn gives e!ect to it.

Viewing history as a literary artifact recognises the importance ofnarrative explanation in our lives as well as in the study of the past, andit ought to liberate historians as we try to narrate the disruptive dis-continuity and chaos of the past for and in the present. This desire is, initself, a product of our own age’s preoccupation with understandingthe nature of our seemingly chaotic lives. Chaos Theory, for example, a1990s methodological innovation, is a new aid to our historical under-standing. Interestingly, one of the leading exponents of Chaos Theorymaintains that its use still requires a narrative to explain the past.23 Thisillustrates how history itself is historical, that is, its methods and con-cepts as well as the debates about its nature are the products of his-torical time periods. In the 1890s American history turned towardsexplaining the peculiarly American origins of the nation’s history, andin the 1950s the strains of the US–Russian Cold War produced aconsensus among historians on the ideological coherence in American

Introduction 15

history in the face of an implacable and potentially divisive enemy. Themillennial rediscovery of the importance of narrative as an access to thesublime and possible worlds of the past is very much the product oftoday and, like all historical understandings, will presumably pass awaywith time. This book and the issues I raise in it are, indeed, very muchproducts of our time.

POSTMODERN NARRATIVES AND HISTORY

The deconstructionist view of history – as a constituted narrative ratherthan the report of an objective empiricist undertaking – results from thewider end-of-century postmodern intellectual context.24 It is a contextthat the French cultural critic Jean-François Lyotard, in his highlyinfluential 1984 book The Postmodern Condition, described as centringon the vexed relationship between the acquisition of what he calledscientific knowledge and the functioning of narrative. In defining narra-tive, Lyotard suggested that it is the characteristic and essential featureof cultural formation and transmission.25 Lyotard agrees with Foucaultthat narrative is about the exercise of power. For Lyotard it is a kind ofself-legitimation whereby constructing it according to a certain set ofsocially accepted rules and practices establishes the speaker’s or writer’sauthority within their society, and acts as a mutual reinforcement ofthat society’s self-identity.26 As a Western cultural practice, history hasbeen challenged by the loss of our self-identity. Meanwhile, historianswho stick to their realist belief in commonsense, science-inspired,objective empiricist paradigm remain inured to what they see as mere‘distractions’ in the pursuit of truthful historical knowledge (eventhough they realise that technical problems with the evidence, socialtheorising, or simple bias may prevent its attainment).

Science, from the eighteenth up to the early twentieth century, hasdepended upon powerful, socially constructed, political and philo-sophical ‘master’ narratives to support, protect and legitimise it – whatLyotard calls meta-narratives. In the epistemological hierarchy the keymaster or meta-narratives were the eighteenth-century Enlightenment(as focused in the upheaval of the French Revolution), promising as itdid human freedom through emancipation from monarchical despot-ism and feudalism, to be followed by the nineteenth-century narrativeof human consciousness leading towards some perfectible future (aselaborated in the philosophy of Hegel). Consequently, Lyotard claimsthat scientific knowledge cannot describe its truth without resortto these other two meta-narratives of emancipation and self-con-sciousness. Science denies narrative as a form of legitimate cognition

16 Introduction

(that is, it is not scientific) while depending on it for its own socialacceptance and intellectual and cultural legitimation.

If by implication, history, like science, is now under challenge today,it is presumed to be partly because of the traumatic events of the twen-tieth century which have meant a loss of confidence in our ability torelate the past or, as Keith Jenkins describes it, ‘the general failure . . .of that experiment in social living which we call modernity’.27 The meta-narrative of scientific objectivity and the unfolding of progress throughour grasp of the past is now under challenge. The rise of fascism, twoworld wars, de-colonisation, seismic technological change, environ-mental and ecological disaster, the information explosion, the growthof exploitative and non-accountable global capitalism, with its com-modification of labour in the ‘developed’ West and the worsening dis-possession of the toiling masses across the undeveloped globe, have allbut destroyed the meta-narratives that legitimised both science and his-tory as foundations of what has been regarded as an inexorable trendtowards individual freedom and the self-conscious improvement of thehuman condition.

As a consequence of all this, at the start of the twenty-first century,narratives both grand and petty, beliefs, attitudes, values, disciplines,societies, and meaning itself, appear to be fractured or fracturing. Thefuture is one of gloomy uncertainty. It now seems quite incredible thatanyone could have ever believed in the hierarchy of master narrativeslike liberalism, science, Marxism, socialism, or a view of history thatemphasised either the discovery of the past as it actually was, or eventhe inevitability of progress. So it is that Lyotard describes the post-modern condition as an incredulity towards meta-narratives. We havenow lost the old, modernist sense of history as the fount of wisdom orteacher of moral or intellectual certainty. What this means is that anystudy of what history is cannot be other than located within its socialand cultural context. History, as a form of literature, is like music,drama and poetry, a cultural practice. As a text or series of texts (evi-dence and interpretations), history can be understood only when it issituated, as the philosopher of postmodern history F.R. Ankersmit saidin the late 1980s, ‘within present day civilisation as a whole’.28 For ourpurposes this means studying both the content of the past and its inter-pretation in its narrative form. As a self-reflexive historian, I definewritten history as a socially constituted narrative representation thatrecognises the ultimate failure of that narrative form to represent eitheraccurately or objectively. We can study the past only by first probing thenature of history as a discipline.

Introduction 17

CONCLUSION

This definition of history, as a literary, cultural practice, places it withinits present postmodern context. From this perspective, the explosion ofwritten history, as well as in its many other forms, will continue to fillthe space provided for it. Historiography well illustrates this eruption inour knowledge of the past, as well as our irruption into it. Not only isthere more history but historians agree on it less.29 That the past is neverfixed is the message of the deconstructive consciousness, whether interms of its epistemology, treatment of evidence, the construction ofexplanations, or the precise nature of our explanatory narrative form.This postmodern or deconstructive history challenges the traditionalparadigm at every turn – hence its description variously as the decon-structionist, deconstructive or linguistic turn. Deconstructionist historytreats the past as a text to be examined for its possibilities of meaning,and above all exposes the spurious methodological aims and assump-tions of modernist historians which incline them towards the ultimateviability of correspondence between evidence and interpretation, result-ing in enough transparency in representation so as to make possibletheir aims of moral detachment, disinterestedness, objectivity, authen-ticity (if not absolute truthfulness) and the objective constitution ofhistorical facts – allowing the sources to speak for themselves. Becausetoday we doubt these empiricist notions of certainty, veracity and asocially and morally independent standpoint, there is no more historyin the traditional realist sense, there are only possible narrative repre-sentations in, and of, the past, and none can claim to know the past as itactually was. It is to this claim that I now turn by addressing the fourkey questions in more detail.

18 Introduction

2 The past in a changing present

INTRODUCTION

Never before has there been such a vast array of methods available withwhich to study the past, such a range of subject-matter and variety ofaudiences, and all to be understood within the broad sense of irony thatseemingly encompasses Western culture today.1 Never before have somany historians also accepted that written history deploys a system oflanguage that is a part of the reality being described – a representationthat is itself a complex cultural as well as a linguistic product. Living aswe do in an age conceived and understood predominantly in terms ofan ironic consciousness, and heavily influenced by the profusion andconfusion of structuralist, post-structuralist, symbolic and anthro-pological models of the relationship between explanation and theory,even the strongest supporters of the traditional empiricist paradigmoccasionally ask how can the reality of the past be known to us – ormore precisely, how accurate can be its representation as a narrative?The debate on the relationship between postmodernity and historycentres on the connection between the empirical and other methods ofunderstanding as used by historians.2

Specifically, the impact of postmodernism on the study of history isseen in the new emphasis placed on its literary or aesthetic aspect, butnot as before only as stylistic presentation, but now as a mode ofexplanation not primarily dependent upon the established empiricistparadigm. Even the staunchest defender of empiricism, Peter Gay, hasnoted that ‘style . . . is worn into the texture of . . . history. Apart from afew mechanical tricks of rhetoric, manner is indissolubly linked tomatter; style shapes, and in turn is shaped by, substance’.3 This shouldbe seen not as subversive but as liberating for the writing of the past.The collapse of the old universal standards upon which modernityas a phase of history was primarily founded – science, liberalism and

Marxism – has meant that history, while it can no more depend onundisputed notions of truth, objectivity and factualism, can speak tonew and even more challenging questions about how we gainknowledge about the past.

THREE APPROACHES TO HISTORICAL KNOWLEDGE

In the introduction, I argued that historians are today addressing fourbasic questions about the method or form of history as well as itssubject-matter or content. The first of those distinct, but inter-related,questions is the big one of whether or not history is an epistemologywith its own rules for gaining and using knowledge. Does history existas a separate empiricist discipline, or is it at best only a branch of theconstructionist social sciences, or possibly a form of literature; or is it sovague an intellectual undertaking that it can be either or both, depend-ing upon the choices made by the individual historian? The answers tothe other three questions, on the treatment of historical evidence, therole of social theory, and narrative as the form of historical expla-nation, animate our answer to this, the big question. In the welter ofhistory today I have already briefly noted three major approaches whichI have characterised as reconstructionism, constructionism and decon-structionism. The reconstructionist, or as it is sometimes called thecontextualist, approach refers to the established consensus or ‘com-monsense’ empiricist tradition handed down from the nineteenth cen-tury. That it actually covers a variety of empiricisms is demonstrated inthe work of hardened reconstructionists such as G.R. Elton, GordonS. Wood, H. Trevor-Roper, Lawrence Stone, John Tosh, GertrudeHimmelfarb, Arthur Marwick, J.H. Hexter and Oscar Handlin, and inthe work of those we might call the practical realists such as PeterNovick, Joyce Appleby, Lynn Hunt, Margaret Jacob, David M.Roberts, Gabrielle Spiegel and Carla Hesse. Both groups constructhistorical explanations around the evidence while maintaining afoundational belief in empiricism and historical meanings ultimatelyderiving from sense experience as mediated by their constructed narra-tives.4 Elton and Marwick are among the most outspoken advocates ofthe modernist ‘craftsman’ approach to historical study, maintainingthat history is still about objective and forensic research into thesources, the reconstructing of the past as it actually happened, and thefreedom of the whole process from ideological contamination and/orthe linguistic a priorism of emplotment and troping.

Constructionism refers to the ‘social theory’ schools that appeal togeneral laws in historical explanation, as exemplified, for example, in

20 The past in a changing present

the French Annalistes, attempt at all-encompassing total explanations,and other sociologically inspired case-study and biographical work ofhistorians like Norbert Elias, Robert Darnton, Marshal Sahlins andAnthony Giddens.5 Modernisation theory is yet another variety of con-structionism which found favour, especially in the USA, in the early1960s. This theory looks to the past for models that might be appliedtoday as a means of studying the present development of the ThirdWorld. The most famous of all constructionist approaches, of course, isthe Marxist/neo-Marxist school as exemplified in the work of EugeneGenovese, George Rudé, Perry Anderson and E.P. Thompson andpolitical scientists who have strayed into history such as Alex Cal-linicos.6 The question usually asked of all varieties of constructionismis how can such history approximate what actually happened in the pastwhen, in e!ect, all it does is generate explanations grounded in con-temporary cultural practices, and hence is ideologically tainted? Thisquestion will remain open for the moment, but I shall return to itbecause it is an open issue that deconstructionist historians mustalso face.

The final group of approaches, loosely defined as deconstructionist,derive their focus from the postmodern historical understanding foundin the work of a growing number of historians and philosophers ofhistory like Hayden White, Dominick LaCapra, David Harlan, AllanMegill, Keith Jenkins, F.R. Ankersmit, Philippe Carrard, Joan W. Scott,Patrick Joyce, Roger Chartier and many other ‘new wave’ intellectualand cultural historians, where the emphasis is placed less on traditionalempiricism or explicit social scientific theorising, than on the relation-ship of form and content (sources and interpretations) and theunavoidable relativism of historical understanding.7 The decon-structionist consciousness accepts that the content of history, like thatof literature, is defined as much by the nature of the language used todescribe and interpret that content as it is by research into the docu-mentary sources. Deconstructionist historians tend to view history andthe past as a complex series of literary products that derive their chainsof meaning(s) or significations from the nature of narrative structure(or forms of representation) as much as from other culturally providedideological factors. Because we historians choose our words with greatcare, it seems wrong to ignore them as a significant part of our attemptto explain the past. I will now outline all three approaches in moredetail before assessing their significance to the writing of history.

The past in a changing present 21

Reconstructionism

The Western tradition of history-writing is built on the correspon-dence theory of empiricism firmly rooted in the belief that truthfulmeaning can be directly inferred from the primary sources. It is fur-ther held that this is enough to constitute history as a separate andindependent epistemology.8 Reconstructionism thereby rests on theassumption that the more carefully we do it, like experienced crafts-men and women, the more accurate we can become, and the closerwe get to fulfilling Leopold von Ranke’s nineteenth-century dictumwie es eigentlich gewesen, or knowing history as it actually happened.The central tenet of this variety of hardened empiricism in historicalstudy is an antipathy to the testing of preconceived theories ofexplanation. Such empiricists verify their knowledge of the past byinsisting that their experience of the real world must be as una!ectedby their perception of it as possible – they remain objective in otherwords. We can gain a useful insight into the conservative heart ofempiricism by reading G.R. Elton’s aptly entitled 1991 book Returnto Essentials. Elton insists that the most valuable aspect of the histor-ian’s work is the ‘rational, independent and impartial investigation’of the documents of the past.9 Arguing that this reliance on com-monsense empiricism does not constitute a theory of knowledge butis history as it should be properly understood, he goes on to dismissrelativism in history – other theories of knowledge – as ‘Ideologicaltheories . . . imposed upon the reconstruction of the past ratherthan derived from it’. For Elton, ideology is the arch-enemy ofempiricism.

In a further rejection of the taint of ideology, bias and the inter-ventionism of the historian, Elton also forcefully rejects the notion thatwriting history may involve a ‘re-enactment in the historian’s mind’. Indisparaging the two most well-known relativist historians, BenedettoCroce and R.G. Collingwood, who in the first half of the twentiethcentury suggested that historians play an active role in constructinghistory by rethinking the past, Elton’s claim that the ‘history of ideas’has now been ‘suddenly promoted from the scullery to the drawing-room’ is not far from the mark.10 Most historians today feel that theycannot ‘do’ history without actively thinking about their role in theprocess of deriving historical knowledge – they do not share Elton’sfaith in empiricism. Indeed, there is a continuing debate (sometimescalled the Historikerstreit, or conflict of the historians) between post-modernist and modernist historians over whether we can ever have agenuine knowledge of the real past, given the opacity and instability of

22 The past in a changing present

language in its constructed narrative form as well as history’s ideo-logical dimension.11

From the Eltonian reconstructionist position, the infection of ideol-ogy produces the greatest of all ills in the fall from the grace of objectiv-ity witnessed in the imposition of the intrusive authorial voice of thehistorian. This can lead only into perspective – a degraded history writ-ten from a particular point of view. The voice of the historian shouldnever drown out the voice of history. For Elton, the impositionalismeither of social theory or ideology is ‘one of the most pernicious com-monplaces of the contemporary analysis’,12 and each generation mustavoid writing history in its own image. With ‘strident’ feminist histor-ians in mind, Elton describes this ‘corruption’ as often the result of‘bigoted idleness’.13 In spite of his bellicosity, Elton raises an importantpoint about whether historians should measure the past according topresent standards of method and/or morality. This is clearly a realproblem if history is assumed to be the objective pursuit of truth. Hisanswer is firmly that it is and that we do history for history’s sake andnot to comment on life today.

Conservative reconstructionist historians are ill at ease with import-ing the discipline of philosophy (usually described as the history ofideas) into their practice. Some (like Elton) are simply anti-theory inany shape, whilst most just oppose the theory or categories of analysisof which they do not personally approve. Elton, for example, rejects notonly the ‘ideological theory’ of Croce and Collingwood, and morerecently the methods advocated by the influential British historian E.H.Carr, but also a range of other theories derived from the social scienceswhich, he claims, tend ‘to arrive at their results by setting up a theor-etical model which they then profess to validate or disprove by an“experimental” application of factual detail’.14 Elton finds Marxism tobe especially pernicious, and gets substantial support from anotherhard-line reconstructionist, Arthur Marwick. In Marwick’s late mod-ernist view, history is not a social science and is, therefore, a non-theoretical exercise. In spite of their shared suspicion of philosophy,their views are supported by a number of philosopher-historians likeChris Lorenz, James Kloppenberg, J.H. Hexter, C. Behan McCullaghand Michael Stanford.

Marwick and Elton argue that history and the social sciences aredistinct because the raw material of history, in the form of unique orsingular documents and relics of the past, precludes the formulation of‘theoretical constructs’, and if the attempt is made, ‘these constructsare nearly always of a more abstract character than the historianwould be prepared to accept’.15 In the early 1990s another sober

The past in a changing present 23

reconstructionist, Lawrence Stone, took up a hard-line position whenhe publicly denounced ‘attacks from extreme relativists from HaydenWhite to Derrida’ on the ‘hard won professional expertise in the studyof evidence that was worked out in the late nineteenth century’.16 His-tory, according to Stone, Elton and Marwick, deals with the historicalconcrete, not the speculative constructions of social scientists and evenless those of deconstructionist philosophers of history and language.Imposing paradigms or models of explanation on the evidence meansthat the past cannot be thought, in practical e!ect, to have existedindependently of the historian who works to understand it. Using the-ory means that we, as historians, impose models of explanation on theevidence of the past which are derived from social sciences, or increas-ingly from other modes of organisation knowledge like structuralism,post-structuralism, anthropology and literary theory. In this sensedeconstructionism, for empiricists, is just one more kind of con-structionist-type imposition on the past. Hard-core reconstructionisthistory is history proper, and proper history has no social theory orphilosophical axes to grind.

Constructionism

Constructionism is essentially a sub-species of reconstructionism. Itgrew in the course of the twentieth century out of the weaknesses of thetraditional reconstructionist paradigm.17 The great complexity and var-iety of constructionism today results from the fact that most historiansrange themselves around the methodological point at which con-structionism branches from reconstructionism. Historians today areprobably more open to new ways of doing history than ever before. Thisbranching begins with the recognition of the frailty of empiricism. Thefirst practitioners of constructionist history in the nineteenth century –Karl Marx, Auguste Comte and Herbert Spencer – were dissatisfiedwith reconstructionism’s simple descriptive narrative of discrete andsingular events. For these nineteenth-century precursors of social the-ory, and subsequently for many others in the twentieth century, historycan explain the past only when the evidence is placed within a pre-existing explanatory framework that allows for the calculation of gen-eral rules of human action. These general rules are revealed as patternsof behaviour, and singular events are seen as part of a discernible pat-tern. In the twentieth century the starting-point of this constructionisthistory was the New History movement of the 1920s, associated withthe French school of historians who gathered around the journalAnnales, and the American New Historians Frederick Jackson Turner,

24 The past in a changing present

Charles Beard, James Harvey Robinson and Vernon L. Parrington. Asthe result of the branching process, the close of the twentieth centuryhas seen an ever-greater variety of ways in which reconstructionism(narrative single event history) and social theory constructionism canbe combined. The richness of constructionist history is witnessed in itsdevelopment in the French Annales school from the work of MarcBloch through to Fernand Braudel, to Emmanuel Le Roy Ladurie andRobert Darnton today, and in the work of anthropologically inspiredhistorians like Natalie Zemon Davis. What is sometimes called culturalMarxism is another example of the development of narrative empiri-cism into a form of constructionism and is well represented in the workof self-styled empiricist Marxist historian E.P. Thompson. For suchpractitioners, constructionist model-making is not taken necessarily toinvolve fitting events into a preconceived pattern. For all these histor-ians, as with those of the Modernisation School, the imposing of anexplanatory framework does not diminish human agency, intentional-ity, or choice in the past, but rather enriches our understanding of it.

As modernist visions of history, reconstructionism and its con-structionist cousin branched not over what is their shared belief in theseparate existence of factual knowledge derived from observable evi-dence, but upon the empiricist claim that it is possible to build highorder and well-justified interpretations upon observable and singularevidence alone. What is challenged by sophisticated constructionism isthe implicit reconstructionist belief that historical investigation canresolve historical issues by evaluating unique events as the peculiarlyhistorical litmus test of knowledge.18 Very few reasonable reconstruc-tionist or constructionist historians today endorse the rigidly conser-vative Elton-Marwick view of history as a purely evidence-basedundertaking that is a non-philosophical and non-theoretical practice.History cannot be written as if it were in some way entirely removedfrom the experience of the present, of our everyday life or the dominantideas within the broader intellectual community. Nor less is it able toavoid explanatory frameworks that must be to some greater or lesserextent culturally provided.

Many historians clustering around the point at which reconstruction-ism branches from constructionism accept that they mediate past real-ity through a complex mixture of professional if not social convention,and shared categories of analysis and conceptualisation if not actualideological positioning. The more ideologically self-conscious socialand cultural historians since the 1950s and 1960s have, however, writtenhistory as an interventionist exercise of dissent and opposition. This isparticularly evident in the interpretations by leftist historian-activists

The past in a changing present 25

like E.P. Thompson, Philip Foner, Christopher Hill, John Saville, MikeDavis, George Rudé, David Roediger, Victor Kiernan, Herbert Gutmanand Raphael Samuel – just a few among many. Although Elton severelydisapproves, the present varieties of social history are evidence thathistory is more and more constructed and written as a form of politicalcommitment to marginalised groups – racial and ethnocultural, gen-dered, class, colonial, sexual and regional. Much social and culturalhistory now being written assumes that the historian’s personal beliefsand commitments cannot be suspended, but that this does not diminishthe value of our historical understanding. It is also more and morewidely recognised that there is, in addition, another important dimen-sion to the writing of history – the form it takes. Notwithstandingclaims by some in the Annales camp to the contrary, even the mostpositivist of constructionist history has to be written as a narrative. Themain point of deconstructionist history is this recognition, that theprimary function of historians, whether they are ostensibly reconstruc-tionist or constructionist, is to narrate a story based upon theirunderstanding of other narratives and their pre-existing interpretations.

This recognition was signalled by a concerned Lawrence Stone first in1979, and again in 1991/92. In his 1979 article ‘The Revival of Narra-tive’, Stone claimed to detect the end of social theory (constructionist)history and, as the article title suggests, the return to an earlier kind ofnarrative-based history (reconstructionism).19 Developments in the nextdecade or so prompted a second foray of 1991/92 in which he character-ised the relationship between ‘History and Post-Modernism’ (the titleof the article) as producing three new threats to history – from lin-guistics, anthropology and the new historicism.20 Despite responsesdefending the new historicist turn in social and cultural history whichnoted that for all practical purposes the events and processes of the pastare indistinguishable from the forms of their documentary representa-tion and the historical discourses that constructed them, Stoneremained convinced that history was in danger of losing sight of theessential character of its elemental empiricist and contextualist founda-tion, as he said, because of ‘the extreme position that there is no realityoutside language’.21

One of Stone’s antagonists, the British social historian Patrick Joyce,claimed that there was a crisis in the history profession and that itcentred on a three-fold consideration: first, that language constitutesmeaning in the social world; second, the historical object of study isalways created by the historian; and last, our access to the past is onlyever through a text – the text as the historian’s written interpretation, oras documentary evidence: diaries, statutes, graveyard memorials, wills,

26 The past in a changing present

films, or whatever. As a result, history is also about the relationshipbetween such texts and our past and present social life as mediatedthrough language-use. Since the late 1970s, literary theory has estab-lished its influence over historians as it has over many other peopleworking in the humanities and social sciences. We historians areincreasingly aware, for example, of the literary rules that govern theproduction of our texts and the nature of historical narrative represen-tation. Leading French historian Roger Chartier concludes that alltexts (whether literary or historical, evidence or interpretation) are bestviewed as the result of a constructed production and reading by thehistorian. They are a representation of the past rather than the objec-tive access to the reality of the past.22 As the historian consumes theevidence of the past, he/she also produces a meaning. How we organise/emplot the evidence creates the past for us and our readers. Thisunderstanding is at the heart of deconstructionist history.

Deconstructionism: narrative and history

Historians of the deconstructionist or linguistic turn, like others awareof the indeterminate character of postmodern society and the self-referential nature of representation, are conscious that the written his-torical narrative is the formal re-presentation of historical content.23

This consciousness has emerged in the last quarter of the twentiethcentury, prompting all historians to think self-consciously about howwe use language – to be particularly aware of the figurative character ofour own narrative as the medium by which we relate the past andwritten history. This means further exploring the idea that our opaquelanguage constitutes and represents rather than transparently corres-ponds to reality, that there is no ultimate knowable historical truth, thatour knowledge of the past is social and perspectival, and that writtenhistory exists within culturally determined power structures. AsChartier has argued, no text, even ‘the most apparently documentary,even the most “objective”, can ever ‘maintain a transparent relationshipwith the reality that it apprehends’.24

The French post-structuralist philosopher and cultural critic JacquesDerrida coined the term deconstruction to challenge the fundamentaltenet of Anglo-American and European philosophy and reconstruc-tionist history: that there is a stable/knowable reality ‘out there’ that wecan access accurately. It is upon such a belief that the basic polaritiesof real–unreal, fact–fiction, truth–untruth, subject–object, andmind–knowledge were established in our culture.25 The thrust ofliterary deconstruction – that there is no certainty of meaning in

The past in a changing present 27

language-based texts because ‘out there’ is always encountered as asocially constructed text – has provoked outrage among empiricistphilosophers and commonsense contextualist historians. The idea thatwe constantly intervene in the real world through language means thatwe cannot achieve a direct representation of reality, and the cor-respondence theory of knowledge collapses. While at one level it mayseem obvious that our world is known to us only through language, andthat language-use makes knowing possible, this is never acknowledgedby inured reconstructionists as central to the writing of history, or if itis noted by a few, it is as just another constraint among many.

As soon as we acknowledge that written history is open rather thanclosed in meaning, as when, for example, the history of imperialism iswritten from a non-European perspective – not recognised as a per-spective at all in the West until the second half of the twentieth cen-tury and the advent of decolonisation – then we come closer to whatpostmodern history means: a recognition of the relativism of meaning,determined by where one stands and the dissolution of source-derivedcertainty in historical representation. But most historians clusteringaround the reconstructionist/constructionist axis still insist on seekingout the past, as opposed to a possible history. They accept evidence asthe essential proof that something discoverable and recoverablehappened in the past, reasoning that the source, if studied appropriately– in its context and/or the application of appropriate models ofexplanation – will reveal the reality behind it. The deconstructionisthistorian, on the other hand, maintains that evidence only signpostspossible realities and possible interpretations because all contexts areinevitably textualised or narrativised or texts within texts. When wehistorians interpret the past we write texts to marshal ideas, to sift andsort evidence, and thereby inevitably and primarily impose a narrativeor textualised shape on the past. The implications of this textual impo-sitionalism are substantial. If deconstructionist historians are correct,and history as knowledge is not discovered, but is produced in andthrough language – as a text – then there can be no reality shorn ofpresupposition, nor the interpretative shaping of historians.26 This isnot a disagreement over historical objectivity, but more one about howthought itself can apprehend the presumed ‘real’ or truthful world ‘outthere’ through an acknowledgement of the ‘varieties of realities’, oreven recognise history’s ultimately meaningless, and therefore open,nature.

28 The past in a changing present

STRUCTURALISM

This basic challenge to empiricism, and particularly its belief in thepower of language to explain through correspondence theory, had itsorigins at the start of the twentieth century, in the broad intellectualenterprise known as structuralism. What we might call orthodox struc-turalism maintains that we perceive and interpret the real worldthrough an innate and pre-existing or a priori mental grid. This gridworks at the deep level of human consciousness and is exhibited in thereal world in many ways, as the structure of grammar, as kinship rela-tionships, myths, even as patterns of food consumption. This meansthat any body of information, like historical data, can be understoodonly through pre-existing or generic mental structures located in themind of the historian. Understanding does not come unsullied with thedata, and the data do not possess inherent and discoverable empiricaltruths, or direct and unmediated correspondences with reality. Thepoint is that structuralism emphasises the formal qualities of what is ana priori internal mental system of understanding, rather than theindependent power of external determinants. As the British Marxistcultural critic Raymond Williams self-reflexively pointed out, althoughthere have been many variations in the use of the term structuralism,the ‘primary emphasis is on deep permanent structures of which theobserved variations of languages and cultures are forms’. The inevitableresult, as he noted, has been a growing ‘rejection of historical (histori-cist) and evolutionary assumptions’ about how we acquire knowledgein the humanities and social sciences.27 It is this structuralist insight thatpost-structuralists, Roland Barthes Michel Foucault and Hayden Whiteuse as springboards for their analyses.

Structuralism has had a profound e!ect on the way in which we thinkabout the past as history, as well as the present and, for that matter, ourfuture. As a theory of how we acquire knowledge, structuralism quicklyplaced the notion of scientific objectivity under sustained pressure asthe relativist basis of knowledge emerged, resulting in the more recentintellectual developments of post-structuralism and new historicism.Its ramifications now cross all fields of knowledge – the natural and lifesciences, the law, anthropology, cosmology, sociology, philosophy, lit-erature and history. As we shall see, structuralism, but particularly itssuccessor post-structuralism, as well as their joint o!-spring new his-toricism, have ultimately all helped to shape the deconstructionistobjections to traditional history.

It was in the study of linguistics that structuralism had its beginnings.Between 1907 and 1911, a Professor of Linguistics at the University of

The past in a changing present 29

Geneva, Ferdinand de Saussure, delivered a series of three lecturecourses. Upon his relatively early death at the age of 56 in 1913, some ofhis friends and colleagues published a synthesis of his lectures andnotes as a book, the Course in General Linguistics.28 In the Course Saus-sure detailed his ideas about the relationship between words and theirsocial meaning. In so doing he produced two arguments that havebecome central to linguistics and to our understanding of the role oflanguage in the creation of all knowledge, not just historical knowledge.

The first is that language operates according to its own rules and isquite unrelated to the ‘real world’, past or present. Saussure explainsthis apparently strange idea through his formulation of the langue andparole – the former being language’s structure, the latter the actualexamples of the system in operation, usually an utterance or expression.Saussure does not see language as just a huge collection of picturesreflecting the reality of things – for example, that the quality or realityof being a horse naturally corresponds to the word horse. In Saussure’sopinion, words do not unproblematically correspond to the things towhich they refer – their referents. In other words, the claim seems to bemade that there is no natural relationship between the word and theworld. The relationship between words and what they signify is, there-fore, arbitrary. Any referentiality we assume in language is the result ofit being fixed by social convention.

Saussure’s second argument follows from this lack of natural cor-respondence of word and world. Words are only ‘signs’ being, in e!ect,defined by their di!erence from other words in a sentence. Signs areconstructed out of two elements – the signifier (the word) and the signi-fied (the concept that the word represents). The structuralist view oflanguage concerns itself only with the structure of the arbitrary connec-tions between signifiers, instead of gazing beyond the language systemat the signified. The important point about the arbitrary signifier–signified relationship is that it is socially produced. Although we alltend to use words as if they were strictly referential, they are of coursebased on conventional social meanings or generally accepted socialvalues. Saussure’s primary insistence on the langue means rejecting thehistorical, or diachronic, dimension of language in favour of the struc-tural, or synchronic, as he calls it. With all this, Saussure created the newscience of signs – semiology or semiotics. The subsequent impact ofSaussure’s work can hardly be exaggerated in the humanities and morewidely, not least in the production of its primary intellectual responsethat is discussed further below, post-structuralism. As the historian Wil-liam Pencak has argued, the study of the past is inherently about thecollection and selection of signs in order to tell a story and construct

30 The past in a changing present

interpretations from signs of real events.29 Arguably, this has beenmanifested in the rise of new empiricism’s e!ort to compromise theunderstanding of ‘the real’ in the only medium we have for such anactivity – language.

We need to understand the implications that this has, particularly forhistory. Because the importance of the sign resides in this arbitraryconnection between signifier and signified, it follows that language isthe complex and determining expression of our experience of life andbeing. We live in a social world of language, and thus language is alwaysfreighted with social meaning and is, as Foucault argues, homologousto the power relationships creating the social structure. It follows thatlanguage, in describing experience, is unavoidably ideological. Ideologymay be defined as a mode of thinking that is in some way or otherrelated to the hierarchies of society and the dispensation of powerwithin it. Consequently, language is never innocent. The definition andmeaning of words/concepts are always connected to the use of power inour society. We shall return to this very important issue again when Ifurther discuss the special contribution of Michel Foucault to thedeconstructionist consciousness.

Structuralism’s notion of the text as a self-su"cient sealed systemmeans that structuralist-inspired literary critics do not comprehendtheir sources – fictional texts – by studying them in their context of reallife. The structuralist literary critic tries to understand them by isolatingtext from context, attempting to work out how the text figures accord-ing to some deep grammatical or syntactic structure. This arcane formof literary criticism is unattractive for most literary critics who prefer torelate their texts to the real world to grasp their meaning. Structuralismin its pure form insists that we have to become detached from thisinvolvement, but this is not possible for historians, dealing as we dowith society. The one implication of this structuralist preoccupationwith the nature of language that is of crucial importance for historiansis the arbitrary nature of signs, which emphasises the problematicnature of language as an e!ective medium of expression and under-standing. If structuralism recognises the importance of language, thenpost-structuralism acknowledges its limitations as a means of under-standing. Accepting the elusive nature of the text as full of gaps,silences and uncertainties of meaning – unfixed and flowing signifiers –it suggests that historical interpretation of texts, like literary criticism,must be indeterminate and that all its readings are more or lessinadequate. This does not, of course, mean that any interpretation is asgood as any other; it simply means that there are no definitive interpret-ations.30 And, of course, it does not stop people (including historians)

The past in a changing present 31

making sense of the everyday world, even though signs are arbitrary. Infact, because signs are culturally determined we learn quickly whatchanges have occurred and we reread them rapidly.

POST-STRUCTURALISM

This view of language as an infinity of free-flowing signifiers that haveno knowable, and therefore no concrete, point of origination, andconsequently no certain end, has been the central concern of JacquesDerrida. To explore this idea of endless chains of signification, Derridaemploys the structuralist concept of di!érance whereby words aredefined by their di!erence to other words, but meaning is continuouslydeferred as every word leads to another in the system of signification.The French critical theorist Roland Barthes took up this post-structuralist idea of endless chains of signification in the late 1960s and1970s, arguing that in consequence knowledge must have many second-order levels of meaning and signification.31 What we have then is afundamental challenge to the correspondence or referential theory ofmeaning.

Worryingly for mainstream reconstructionist/constructionist histor-ians, if language is uncertain, then our knowledge we gain through itmust be equally indeterminate. This means that it is impossible to con-struct truthful narratives as historical explanations. Despite Derrida’sand Barthes’ post-structuralist argument that meaning is just a streamof signifiers, most historians still insist on the eccentric practice ofreading texts (historical sources and narratives) to locate the truth.They do this because they still believe in the commonsense reconstruc-tionist notion that there is a referent for each word and that con-sequently there is some external presence to the text as evidence that wecan be sure of. Many historians still continue the search for the realhistorical past as it once existed and which they believe can be truthfullyrecovered like treasure trove hauled from the sea bed or a fire re-ignitedfrom the ashes.

The question is what can historians do when confronted by theseissues? Most simply do not think about them. Entertaining doubtsabout the veracity of language leads to a critical process that requiresunravelling the style and figurative dimensions of texts. The vast major-ity of historians are just not interested in that undertaking. They arguethat they do not wish to study the literary form of their own discourse –the books they write. This is the view that the Dutch history theoristFrank R. Ankersmit has been confronting for the past twenty-five yearsin a series of influential texts.32 If we accept that the evidence of what

32 The past in a changing present

happened in the past does not unproblematically determine the mean-ing of the past, but that the historian’s language is essential to thatprocess of meaning creation, then plainly the history we create has to besubject to the language decisions made by the historian. This does notmean we cease to be empirical or rational as historians, but what it doesmean is that we need to be not just aware of the linguistic turn and thentry to circumvent it as new empiricists try to do.

I am suggesting it is not viable for the historian to imagine they canflee from narrative into the past. But, being an anti-representationalistor deconstructionist historian is not the same as being anti-realist. Thedeconstructionist point is di!erent. When we make history, that is whenwe construct narratives, we remain in touch with past reality but, asAnkersmit points out, it is not outside the single statement of justifiedbelief. Ankersmit’s representationalism suggests that what is in factimportant about history is its ontological status as a type of literature.While the substance or content of the past remains our reference point,how we assemble our interpretation of its meaning or shape its form isas much about inscription as it is empiricism and analysis.

However, this issue of relating form to content to context quicklybecame a part of the new historicist movement which emerged in thelast twenty years of the twentieth century.33 Not of initial concern tohistorians, new historicism emerged as a type of literary criticism in theUnited States in the early 1980s.34 Taking its intellectual cue(s) from avariety of post-structuralist literary critics and postmodernist thinkers,new historicism challenged literature’s established disciplinary bound-aries while ventilating further doubts about language as a transparentmedium able to generate meaning by corresponding to the social world,past and present – a debate that paralleled and fuelled doubts about therepresentational power of written history.

NEW HISTORICISM

By the early 2000s new historicism had moved beyond literary criticismto assume the proportions of a much wider cultural analysis. AsHayden White has suggested, new historicism was initially little morethan ‘an attempt to restore a historical dimension to . . . literarystudies’,35 to relocate literary works within their historical context – tounderstand poems, novels and plays as texts not simply in their struc-turalist relation to each other, but also in their associative connectionswith the institutions of society and the historical events that mighthave influenced their production: the relationship of text to context.As a cultural analysis, new historicism was yet another twist in the

The past in a changing present 33

continuing exploration of the socially constructed relationship betweenthe knower and the known, between evidence, proof and truth. Forworried historians like Lawrence Stone, new historicism is a threat tothe traditional study of the past because it deals with political andsocial practices as cultural scripts, or language systems or codes, withhistory emptied of its association with past reality36

It is important for historians to understand what new historicism issaying because, like deconstructive history, it is built on assumptionsthat directly challenge the empiricist paradigm37 First, our descriptionsof real historical events like fictional ones can at best only be represen-tations, or events under description, because there is no direct way inwhich historians can acquire first-hand historical knowledge. Second,history as a literary form is about the unique and contingent eventand the real nature of causality must consequently always remainunresolved. Third, historians ought to recognise the overlapping ofhistorical events and their interpretation – not just that the writtenhistory of one generation becomes the primary sources of the next, butthat the historical text itself exists intertextually within the broadersocial and political structures of any epoch. Finally, new historicistthinking suggests that our evidence and the written discourse we pro-duce in interpreting it are time and place specific – there are no uni-versal historical truths to be discovered or transcendental values to beelaborated. These seemingly innocuous assumptions undermine the twomainstream approaches to history because they undercut reconstruc-tionist foundational beliefs in an accurately discernible reality ‘outthere’, and in the fact that we can verify constructionist theories ofexplanation through empirical testing.

Consequently, the distinction between cultural history and other lit-erary disciplines has disappeared under new historicist thinking aboutthe conventions underpinning the representation of factual as well asfictional texts. This opening up of historical analysis to rhetorical inter-rogation is at the heart of deconstructionist history, which recognisesno practical distinction between history proper and the philosophy ofhistory where that includes analysis of the form of history’s writtenform. What this means is that the analysis of emplotment and style,more usually applied to fictional literature, is also basic to the under-standing of all types of historical text, including sources.38 The narra-tive form of explanation is now redeemed as a central feature of thehistorical enterprise, and the notional distinction between historicaland literary language disappears.

Deconstructionist history’s reading of its sources elevates their form,in the shape of their narrative structure, use of metaphor, style,

34 The past in a changing present

emplotment, and so forth, to the same level of interpretative analyticalsignificance as that of content. Form must be acknowledged likeempirically derived content in the production and conveyance of mean-ing and knowledge. This does not mean that the content of the past issecondary or unimportant. What it does mean is that, as the culturalcritic Raymond Williams declares, we need to examine the forms ofcontent and ‘the content of forms, as integral processes’39 Becausedeconstructionist history no longer demarcates literary texts asqualitatively or categorically di!erent from either historical sources orinterpretations, there is no need for a privileged hierarchy between thehistorian’s critical study of the sources and a self-e!acing role for lan-guage and the narrative ordering of the data. The deconstruction ofhistory means no longer repressing the importance of writing history or,more radically, being willing to view the past as well as our existence inthe present as texts to be read.

DECONSTRUCTIONIST HISTORY: FOUCAULT AND WHITE

For reconstructionists, their version of history centres on the nature ofevidence as the key to the accurate recovery of the past. The failings ofthe correspondence theory of truth, the imposition of theoretical struc-tures, the indeterminacy of language or debates over the nature of real-ity are not primary concerns. Two individuals who have challengedmainstreamers to address these perspectives, however, are the Frenchphilosopher and historian of sexuality Michel Foucault, and the Amer-ican historian of the Renaissance and philosopher of history HaydenWhite. Both have addressed the representational function of languagein the production of historical knowledge, and specifically the relation-ship between historical ‘discourse’ and past and present culturalchange. For my purposes, historical discourse is defined as a sharedlanguage-use where meaning derives not directly from the intentionalityof the speaker/writer as either historical actor or historian, nor solely inrespect of the content of what is said or written, but from the formalstructure and context in which the utterance or text is delivered orlocated.40 Taking this definition with its emphasis on history’s socialcontext, both Foucault and White have stressed the capricious nature ofhistorical discourse that results from the arbitrary signifier–signified–referent relationship, and the consequent unsettled social world of thepast and the present.

Foucault especially has acknowledged the post-structuralist ques-tion-mark over narrative’s failure to encompass any genuinecorrespondence to, or reflection of, past reality. The traditional or

The past in a changing present 35

commonsense reconstructionist search for historical origins is not apart of his project. He is a historian who does not believe in the empiri-cist notion of causality. While this lack of belief in itself clearly castshim beyond the bounds of mainstream history, he further compoundshis sinning by both placing the individual historian at the very centre ofthe process of constituting historical knowledge, while at the same timequestioning the centrality of the author as the generator of meaning.His definition of historical truth depends on the agreement betweenhistorians as to what constitutes the historically real – summarised byFoucault as the will to knowledge. For Foucault, knowledge and dis-course are interchangeable, and because both are grounded in the cul-tural practices of society, they are inextricably related to the exercise ofpower both intellectual and material. The central claim of conservativereconstructionists – that the history they write is the discovery of theverifiable truth of the past – Foucault rejects as naïve, or, what is worse,as the perpetuation of a monstrous myth.

In assuming that written history is essentially a form of literature,Hayden White also addresses the issue of history as an epistemologyresting on the distinction already noted between the past and history.Because for White we can never know the story of the past as it actuallywas, it means that there can be no historiographically uncontaminatedpast – the past exists for us only as it is written up by historians. Historydoes not pre-exist in any body of facts that will allow unmediated accessto the real past. History, as opposed to the past, is a literary creationbecause it is always interpreted through textualised relics which them-selves are only to be understood through layers of interpretation as thehistorian’s facts. Because the facts never arrange themselves auton-omously to yield meaning, White argues that it is the function of thehistorian to impose a meaning through the organisation of the dataas a narrative. This necessitates troping and emplotments. This is thepoint at which many mainstream historians reject what they see asWhite’s cutting adrift of history from its factual anchor. White, theyclaim, relativises the past in the light of his now famous post-struc-turalist-inspired, anti-representational and anti-empiricist suggestionthat

historical narratives . . . are verbal fictions, the contents of which areas much invented as found and the forms of which have more incommon with their counterparts in literature than they have withthose in the sciences.41

For White, the historian’s interpretation means selecting from the evi-dence that which is significant and that which, when strung together,

36 The past in a changing present

produces a meaningful explanation, or, as he would say, anemplotment.

The actual mechanism for relating the evidence to the context neces-sitates us formally employing strategies of explanation founded on thetropes (the figures of speech I have already mentioned: metaphor,metonymy, synecdoche and irony), emplotments (the four primary onesof romance, tragedy, comedy and satire), and other strategies ofexplanation which he calls formal arguments (formist, mechanist,organicist and contextualist), as well as explanations through theideological commitment of the author/historian (anarchist, radical,conservative and liberal). These aspects of historical explanation will beexamined in more detail when I discuss White’s model of historicalexplanation in Chapter 8. The important point for now is that the pro-cess of historical explanation for both White and Foucault is one ofliterary e!ect, rather than literal meaning. Historical explanationultimately relies upon the use of the tropes that we all use to expresswhole–part (and the reverse) relationships, and which I have alreadydesignated as the troping process. As French historian Philippe Carrardclaims, historians can try to eliminate such literary devices, but writing‘without turning to tropes is not a simple task, even for scholars who. . . have been trained to do so through such demanding exercises asdissertations historiques’.42 As we shall see, metaphor, the founding trope,as well as its successive refinements in the form of metonymy andsynecdoche, are basic to the constitution of narrative explanationsand the human process of understanding, experiencing and explainingsocial change. Like people in the past (and in the present and future),because we historians cannot escape figuration in narrative we shouldunderstand the nature of its representational character.

CONCLUSION

At the start of this chapter I posed the question: why is it that historykeeps on changing? The deconstructionist answer ought now to beclearer. It changes for two reasons. The first is the postmodern condi-tion in which we live and which confronts the inadequacy of the mod-ernist empirical method; the second, flowing directly from this, is therealisation that history is a constituted narrative discourse written bythe historian in the here and now. History always comes to us at manyremoves from the actuality it claims to represent. Every historical inter-pretation is just one more in a long chain of interpretations, each oneusually claiming to be closer to the reality of the past, but each onemerely another reinscription of the same events, with each successive

The past in a changing present 37

description being the product of the historian’s imposition at the levelsof the trope, emplotment, argument and ideology. No amount of train-ing in the forensic skills of dissecting the sources can eliminate theunavoidable process whereby the historical work is as much invented asfound. History is not a separate epistemology, but as a form of expla-nation is a plausible literature. The fact that the historical narrative isalways figurative confounds the empirical insistence on history as arealistic reconstruction or representation of what happened by cor-respondence to facts. While this is at the centre of deconstructionisthistory, it remains unacceptable not only to the conservative minorityof reconstructionist historians but to all mainstreamers who refuse toslip the anchor of empiricism. It is necessary, therefore, to examine theirvision of the historical enterprise before we can move to my criticismand the implications of the deconstructionive consciousness for thewriting of history.

38 The past in a changing present

3 History asreconstruction/construction

INTRODUCTION

As I have attempted to show, although most historians found in eitherof the two main tendencies agree that history is presented as a writteninterpretative narrative, they assume that it corresponds to what actu-ally happened because of the research carefully undertaken in thesources. They perform this research task believing in the ideal of objec-tivity and attempt to produce interpretations through a value-freeinductive and/or deductive method, and finally reach what is for themconvincing historical explanations. In sum, their interpretation pos-sesses referentiality and corresponds to the truth. What unifies themajority of historians is this general commitment to an evidence-basedmethodology which, by following basic inferential ‘rules of evidence’, ispresumed to produce specific interpretations that allow the reconstruc-tion/construction of the past close to the truth of what it was about. Ishall now review this complex approach to acquiring historical knowl-edge prior to examining the issues raised for deconstructionist history.

EPISTEMOLOGY

Reconstructionism and its positivist model-making derivative, con-structionist history, rely ultimately on a shared belief in both the epi-stemological integrity of empiricism and a recoverable past reality ‘outthere’. One recent commentator claims that, ‘in contrast to radical scep-ticism’, as a historian he adopts ‘the realist position that it is possibleby means of a resort to the tried and tested “historical discourse of theproof” (both theoretical and evidential in character) truthfully toreconstruct the past’.1 Reconstructionists, by assuming an anti-hypothetical and value-neutral empirical method (akin to a positivist orscientific methodology), believe, as that quotation evidences, that they

really can explain the past with a substantial claim to accuracy andtruthfulness. The mainstream reconstructionist philosopher of historyC. Behan McCullagh provides an exceptionally clear argument as to theobjectives of reconstructionists and maintains that most try to discoverwhat actually happened in the past, which explains.

why they pay such attention to the accuracy of their observationsof evidence and to the adequacy of their inferences from it, andwhy they refuse to put forward any descriptions of the past forwhich there is not good evidence. If the pursuit of truth wereabandoned as the goal of historical inquiry, then the main reasonfor insisting upon present standards of historical criticism woulddisappear.2

McCullagh concludes that

although historical descriptions cannot be proved true beyond allpossibility of error, they can often be proved probably true, givenempiricist assumptions. Assuming that an historian’s perceptions ofdata are very probably accurate, that his general knowledge andother beliefs are very probably true, and that his forms of inferenceare generally reliable, one can rationally infer the probable truth ofmany historical descriptions.3

Without this belief in the reliability of a historical description asinferred from the available evidence, we could never claim that historyexists as a distinct epistemology. To the extent that we believe in in-ductive inference, we believe in the truth of historical knowledge. ForMcCullagh, calling the interpretation of a text correct ‘is to say it wouldbe accepted as the meaning of the text by the majority of educatedspeakers of the language in which it is written’. These educatedspeakers would, of course, be aware of the ‘literary and historical con-texts relevant to its subject matter and the intentions of its author’.4 Ashe neatly concludes, summarising the philosophical underpinning ofthe mainstream position, ‘to pursue reliable historical descriptions is topursue true ones’.5 The only ground for doubting this logic is if we denythe essential nature of empiricism, or if the historian’s ‘forms of infer-ence’ are in some way flawed. McCullagh is convinced that the ‘objec-tive understanding of historical texts is possible’, that understanding is‘rationally justifiable’ and that meaning thus derived should be acceptedas correct.6

For most of the twentieth century this way of acquiring knowledgehas constituted a consensus founded on the six key principles ofempiricism:

40 History as reconstruction/construction

• The past (like the present) is real and ‘truth’ corresponds to thatreality through the mechanism of referentiality and inference – thediscovery of facts in the evidence.

• For reconstructionists, facts normally precede interpretation,although constructionists argue that inductive reasoning cannotoperate independently of the deduction of generalised explanations.

• There is a clear division between fact and value.

• History and fiction are not the same.

• There is a division between the knower and that which is known.

• Truth is not perspectival.7

The essential feature of historical truth for all mainstream empiricistsresides in the first principle: that a single historical description, asopposed to an interpretation based on several related descriptions, maybe regarded as true so long as it corresponds to, or resembles, one ormore of a set of truth conditions. This means that we can believe ahistorical description is true if it accords preferably to several knowncriteria or at worst a single criterion of correspondence or referentiality.The criteria of correspondence are usually found through comparingpieces of primary evidence or, less convincingly, the descriptions ofother historians which constitute our secondary evidence. Truthful his-torical descriptions may be taken to depend upon one or more of threekinds of inference: first, that most favoured by reconstructionist/constructionist historians and what I will label as the hypothesis–deduc-tion–data–induction method, or the loop of explanation and evidence;second, statistical probability; and finally, the deconstructive notion ofhistorical justifications derived and implied by our narratives.

McCullagh’s concern, namely the extent to which historians canaccurately recover and represent the past, has been reformulated by theself-proclaimed practical realist historians Joyce Appleby, Lynn Huntand Margaret Jacob in their collaborative and provocatively entitledbook Telling the Truth About History (1994) in which they, like McCul-lagh, promote and defend the correspondence theory of historicalexplanation. In their collective view the debate on the relationshipbetween postmodernism and history comes down to how we bridge thegap between the records of the past, and their narrative interpretationby the historian. As moderates in this debate, they willingly acknowl-edge as a truism that ‘the past only dimly corresponds to what thehistorians say about it’. While they accept the basic reconstructionisttenet of the existence of a discoverable historical truth ‘out there’, aspractical realists they happily concede ‘the tentativeness and imperfec-tions of the historians’ accounts’. Naturally, their commitment to the

History as reconstruction/construction 41

correspondence theory requires that they insist that this does not ‘causethem to give up the e!ort to aim for accuracy and completeness and tojudge historical accounts on the basis of those criteria’. They contrasttheir pragmatism with that of the ‘anti-realists’ or ‘relativists’, whomthey suggest believe ‘any kind of correspondence is impossible’ betweenevidence and written narrative interpretation.8

Because Appleby, Hunt and Jacob have become the most widely readproponents of what we might call the moderate wing of the recon-structionist position, we should take seriously their attraction to‘reconstructing what appears in the mind when it contemplates thepast’. It is appropriate then that we should isolate the six key principlesupon which this reconstructionist/constructionist view of history isfounded – what Appleby, Hunt and Jacob describe as the translation ofwords ‘from the documents into a story that seeks to be faithful to thepast’ and which constitutes ‘the historians’ particular struggle withtruth’.9

This classic, six-point, historical method assumes that advances inthe techniques used to study and draw inductive and deductive infer-ences from the evidence will generate ever more truthful historicalinterpretations. This assumption is summarised in Arthur Marwick’sclaim that ‘the powerful shoulders of our illustrious predecessors [arethere] for us to stand on’, and as a result there is an ‘absolute advance inthe quality, the “truthfulness” of history’.10 Marwick is convinced thathistory is primarily about ‘finding things out, and solving problems,rather than about spinning narratives or telling stories’. He insists that‘History is a human activity carried out by an organised corps of fall-ible human beings [acting] in accordance with strict methods and prin-ciples, empowered to make choices in the language they use . . . andknown as historians.’11 Marwick, like Elton, rejects White’s andFoucault’s claim for the inevitable impositionalism of the historian.Although Marwick accepts the notion of history as a discipline in thesense of a profession, he will not agree that the profession is disciplinedby power relations to say and do certain things. He certainly disagreeswith White’s view that history has been domesticated by ideologies ofall kinds from the nineteenth century onwards, and that deconstructionhas reinvigorated the past by acknowledging its possibilities ratherthan truths to be found. But it would be unfair to suggest Elton andMarwick are the only reconstructionists around. While it is true thatinsistence upon the primacy of reference above all other things, isreconstructionism’s most obvious characteristic, there are many otherhistorians who also emphasise this to the exclusion of virtually every-thing else. Jenkins and Munslow have addressed the assumptions of this

42 History as reconstruction/construction

group of historians at length recently.12 They point, for example, to theBritish social historian Edward Royle’s text Modern Britain: A SocialHistory, 1750–1997 ([1987] 1998) as being emblematic of what theyrefer to as the reconstructionist genre of history.13 The epistemologicalassumption of Royle in what is intended to be a student friendly surveyis revealed in the structure of the text, which is chronological withinthemes. Comprehensiveness is a key feature with the unstated assump-tion that this is really how the social history of Britain during thoseyears really was. Like all reconstructionists Royle o!ers the interpret-ation first, then the evidence in support of it, deploying an authoritativetone throughout. Royle is simply telling the reader what happened and,in so doing, is revealing the story of British social history. The linkbetween reference, explanation, meaning and truth is epistemologicallyunproblematic. While carving one’s way through the archive is agreedto be a highly complex activity, this is not, of course, what makes for thetruth of history. It allows us to understand accurately what happened.But turning reference of what happened into history is not simply anact of (referential) reconstruction. As a textual representation of thepast, history is relative to all kinds of highly complex literary acts anddecisions. No sense of this emerges from reconstructionist history.

However, it would be completely wrong and unfair to suggest thatreconstructionists are unaware that history is about debates over mean-ing. While it is clear that knowing what happened will, so they believe,give us the story, nevertheless history, as they acknowledge, is alwaysabout interpretation. This is exemplified in Michael A.R. Graves’Elizabethan Parliaments, 1559–1601 ([1987] 1996).14 He notes severalschools of interpretation as to the meaning of late sixteenth-centurynational politics. But, as a reconstructionist, he remains tied to thenotion that only fresh evidence will ultimately be the judge of whichinterpretation is ‘right’, that is, best supported by the available evidence.None of this casts doubt upon the knowability of the meaning ofevidence and its translatability into history.

Most mainstream historians start by rejecting what they characteriseas a relativist view of historical knowledge. They agree that the pastonce existed and that the human mind is quite capable of producingstatements about it that are close enough to reality for most practicalpurposes. Behind this practical-realist empiricist approach is the beliefthat the truth once existed and is now discoverable because the eventsand actions that have occurred correspond to the evidence. We can,therefore, be fully justified in making factual statements describingthat correspondence, and any provisionality of historical interpretationsimply means every interpretation is just one more further genuine

History as reconstruction/construction 43

attempt to get closer to the truth – standing on Marwick’s shoulders.The bases for historical knowledge are events and actions representedas empirical facts. This view rejects the deconstructionist position thatfacts are narrativised texts and always, therefore, cloudy, obscure andultimately impenetrable.

Central to empiricism is the assumption that historians, likescientists, search for the truth. For historians this is the assumption ofreferentiality rather than a truth-e!ect. The inheritance of the earlyseventeenth-century English historian and founder of the scientificmethod Francis Bacon remains today in his development of the pri-mary historical method – inductive inference. This method derives his-torical meaning by drawing un-biased inferences from the detailedevidence of individual examples. Baconian-inspired inductivismreached its high tide in the years from the late 1950s to the 1980s in thework of English historians like Hugh Trevor-Roper (in Religion, theReformation, and Social Change, 1967) and G.R. Elton (in England,1200–1640, 1969), and Americans Oscar Handlin (Boston’s Immigrants,1959), Gertrude Himmelfarb (The Idea of Poverty: England in the EarlyIndustrial Age, 1984) and J.H. Hexter (Reappraisals in History, 1969), aswell as philosophers of history like Quentin Skinner (Machiavelli,1981). All accepted the commonsense referentiality of language andforcefully rejected the impositionalism of a priori conceptualisationsarrived at by deductive inference.

Unlike social scientists, reconstructionist historians do not proposegeneral theories, or operate working hypotheses that they then proceedto ‘prove’ by the derivation of facts through empirical research. In-ductive inference requires that theories of explanation emerge from thediscovery of the evidence which translates into meaningful facts afterbeing placed in its historical context. Having said this, as Alex Cal-linicos points out, in historical explanation today deductively inferredtheories are invariably employed, either consciously or unconsciously. Itis impossible, even for the most stainless of reconstructionist historians,to approach their evidence innocent of presupposition, since priorassumptions may, wittingly or unwittingly, await confirmation or denialthrough research – examples of the a priorism so distasteful to Eltonand Marwick. In practice, deduction and induction merge as the pro-cess of historical reconstruction translates slowly into construction andback again. The more epistemologically self-conscious we have becomeabout the possibilities in the theories we employ, and the philosophyof history to which we adhere, helps explain why our constructionisthistory has become ever more complex in the last twenty years.

As Appleby, Hunt and Jacob have demonstrated, few historians

44 History as reconstruction/construction

today deploy a pure form of inductive analysis reliant solely upon acommonsense interpretation of events that assumes, almost as an after-thought, the use of a transparent and unproblematic narrative medium.Nevertheless, most reconstructionists continue to insist that they justifytheir inductive inferences – and therefore sustain the epistemologicalintegrity of the discipline – through the direct observation of the evi-dence of the past. The data thus observed/discovered determine theinductive interpretation, regardless of whether that interpretation dif-fers from that which may be currently dominant among their peers.Consistency, coherence and correspondence with the observable factsremain the watchwords of reconstructionist history now as they havebeen for the past century.

EVIDENCE

In his defence of induction as the historical method, Elton insists that‘history should not be regarded as merely a form of some other intel-lectual enterprise: it has its own operating rules, its own independentfunction, and its own contribution to make to the intellectual and sociallife of mankind’.15 Inductivist historical knowledge, Elton maintains,derives from the authority of the available and validated sources. But asBritish historian John Tosh has said, interpretation of the evidencecannot literally generate a meaning, without ‘a command of the histor-ical context’ which will reveal that to which the evidence corresponds.16

Reconstructionist historians cannot understand the past by only con-sulting the textual evidence. They must place it within the broaderframework of which they are aware, the context, in order to reconstructthe past as it really was. Contextualisation is not the same as configur-ation or emplotment, the latter two being the active product of thehistorian, unlike the former, which reconstructionists assume to bemerely the setting of the scene, the laying out of the adjacent pieces ofevidence, the other pieces in the jigsaw.

Scrupulous attention to the evidence is the bedrock of the six prin-ciples. We could do worse than consult G.R. Elton on the importanceof these principles in the processing of evidence for the reconstruction-ist historian. As he says:

We are looking for a way to ground historical reconstruction insomething that o!ers a measure of independent security – indepen-dent of the historian, independent of the concerns of his day,independent of the social and political conditions imposed on him.And the obvious answer to this quest, as it has always been and must

History as reconstruction/construction 45

continue to be, lies in the sources he has at his disposal. Ad fontesremains the necessary war cry. For the historian the reality – yes, thetruth – of the past exists in materials of various kinds, produced bythat past at the time that it occurred and left behind by its testimony.Historical evidence is not created by the historian, and little of itwas deliberately created for him; it is simply that deposit of pasthappenings that still exist to be looked at.17

Elton thus summarises the traditional key assumptions upon whichevidence is treated and interpreted. This can be undertaken only bythe professionally trained historian, which means the independentlyminded and judicial historian. This training is a mixture of languageskills, a broad knowledge of the context, a deep knowledge of theextant interpretations within the field, and a close understanding ofthe nature of the primary sources which allows comparison and veri-fication. But Elton is at pains to point out that the understanding ofthe evidence is not the same as saying that it is ‘processed throughthe historian’s personal mind’. Instead historians must all ask thesame questions of the evidence – who created it, for what purposes,and how did they create it? ‘Which is to say [that these fundamentalquestions we put to the evidence are] independent of the concerns ofthe questioner and focused entirely on the concerns of the originalcreators.’18 The point is to divorce the historian from the past – notonly to get rid of hindsight, but to avoid writing history from theperspective of the present. The historian’s personal preferences,whether in terms of bias of method, or ideology (or both), must beavoided.

Firm in his belief in the historical method of inductive inference,Elton insists that it subjects ‘every paradigm on o!er to a scepticalquestioning in the light of discoverable detail’. Unlike, say, literature,historians do not have the same free will to put forward interpretationsconstrained only by our imagination. We cannot invent details just tomake our story more convincing. For Elton, reconstructionist history isthus neither science nor art:

For it cannot expect to arrive at knowledge testable by falsification(the secret of a science), nor can it manipulate its subject matter so asto produce morally or aesthetically satisfying results (the character-istic of an art). In short, history is a study di!erent from any otherand governed by rules peculiar to itself.19

It is the study of evidence that not only makes history epistemologicallyindependent but, more significantly, makes it capable of reconstructing

46 History as reconstruction/construction

the past as it actually happened, and without any imposition from thehistorian.

As a result of the practice of objective inductive inference there isalways daylight between fact and value in the study of the traces of thepast. This means never breaching the divide between the knower andthe known when asking questions of the evidence. Questions should beseparated from prior knowledge so that the evidence cannot be skewedtowards an answer already lurking in the recesses of the historian’sbrain. This fences o! the deductive method for Elton. Neither beggingthe question nor begging the answer has a place in reconstructing thepast. In a felicitous phrase Elton summarises this as meaning ‘onesolicits questions from the evidence [for it is] wrong to start with exactquestions carrying in-built answers’.20 For example, it would be badpractice to interpret the late nineteenth-century American economicadvance across the Pacific as a salt-water economic imperialism, whileat the same time asking what social class or classes benefited. This is avalue-laden question that assumes the existence of classes and which,therefore, begs answers. It would be better to inquire of the evidence ofeconomic expansion what was its particular character in comparison toother periods of economic growth, and did the process benefit a par-ticular group, if any? Di!erent forms of questions produce di!erentanswers.

Keeping an open mind about the past assumes that history and fic-tion are not the same and that truth is not perspectival. The applicationof the basic principles of reconstructionist historical analysis will pro-duce conclusions about the past that, though often incomplete or tenta-tive, will serve the social function of keeping the truthful social, politicaland economic memory. To fall below the exacting professional stand-ards set for the reconstruction of the past is, according to Elton, to leave‘the task of telling about the past to the untrained and largely ignorant– to the writers of fiction, avowed or disguised, to the makers of films,to the journalists and speculators of the pen’.21 The distinction betweenhistory and fiction resides in the professionalism of the historian asmuch as in the constraint to recount what actually happened ratherthan invent it. As Michael Stanford has argued, ‘a historical factaccords with a judgement about the past in which historians agree’.22

Stanford points to the di!erence between interpretations and facts – theformer produce no consensus among historians; the latter do. Withoutthis dependence on facticity history cannot exist.

In the last few years, however, a new moderate or practical realistconsensus has dominated reconstructionist/constructionist historicalscholarship. The American intellectual historian David Hollinger

History as reconstruction/construction 47

summarised this consensus when he argued that the historian’s pre-conceptions are often what makes historical interpretation possible.23

This thought, from nearly twenty years ago, still resonates today. This isnot to dispute the evidence, but to acknowledge the murky turning ofevidence into facts (contextualisation) to produce an interpretation.When historians attempt to reconstruct the past by studying its evi-dence – Hollinger’s process of Quellenkritik (the critical examination ofdocumentary sources) – the historian cannot be as isolated from thereconstruction process as conservative empiricist reconstructionists likeElton would have us believe.

The clearest case of impositionalism is in the application ofexplanatory theories to the experience of the past – the deductivemethod as employed by constructionist historians. Deductive reason-ing assumes that knowledge is derived from premises tested by obser-vation. Contemplating this process horrifies hard-core reconstruction-ists like Elton. What he calls ‘interpretative and ideological theory’arises from the ambition to ‘destroy the reality of the past as it hadpreviously emerged from a study of that past’s relics’, and he concludesthat ‘all do equal harm to the independent understanding of thepast’.24 We must, therefore, now examine the role of theory in writingthe past.

THEORIES OF HISTORY: CONSTRUCTING THE PAST

Constructionism covers a variety of impositionalist approaches to thestudy of the past, but all share the reconstructionist’s belief that ourhistorical knowledge corresponds to the reality being studied. Bothmainstreams doubt what they view as the proto-deconstructionistapproach to history exemplified by R.G. Collingwood, but are happierto accept the position of E.H. Carr who typifies the relativist judgementwhich insists that they are facts because the historian has selected themfor inquiry, what Carr calls historian’s facts. It follows, much to theannoyance of Elton, that objectivity is now impossible to attain. Carr’sview of history is that it is concerned with the relationship between theindividual and the general and, as a historian, ‘you can no more sep-arate them, or give precedence to one over the other, than you canseparate fact and interpretation’.25 For Carr, echoing Collingwood’sgeneral position:

the facts of history never come to us ‘pure’ since they do not andcannot exist in a pure form: they are always refracted through themind of the recorder. It follows that when we take up a work of

48 History as reconstruction/construction

history, our first concern should not be with the facts which itcontains but with the historian who wrote it.26

This position has a powerful appeal to constructionist historians.Callinicos, defending the constructionist historian, claims he/she goesabout his/her business by inferring answers from the questions he/sheputs to the evidence and not from the sources which cannot speak forthemselves. Now this may sound reasonable enough, but at the end ofthe day Callinicos ends up in a position similar to Elton’s as they bothpresume to solicit questions rather than answers from the evidence. Ofcourse the di!erence is that Callinicos insists that facts emerge fromanalysis, not analysis from facts. According to Elton, Marxism, as themost well-known form of constructionism, chooses to view historicalreality as being ordered by a bastard version of a so-called covering law.A covering law ascribes causation in history and it is derived fromdeductive inference. An explanation of an event or particular action ismade (deduced) in terms of an invariable law of nature or humanbehaviour. Elton’s rejection of covering laws follows on from his beliefthat historical explanation requires understanding the motivations,goals, values and information available to historical agents, all of whichconstitute their individual intentions and cannot be subsumed underuniversal explanations of behaviour. The divergent views of the ardentreconstructionist G.R. Elton, and the equally impassioned Marxistconstructionist Alex Callinicos, reveal the chasm in the non-deconstructionist historical profession between the extremes ofpositivism and empiricism: what Peter Burke calls theorists andhistorians.27

When social theory historians write history they set about retellingand narrating the lives, intentions and events of the past with models ofexplanation already in their minds – gender, race, class, or whatever. Ofcourse, they usually stress that they are not slaves to proving the accur-acy of one over-arching theory of social action or philosophy of his-tory, unless they are overtly committed to a certain perspective as an actof faith. Instead they maintain that their models are no more than‘concepts’ – though often highly complex in construction – that emergefrom the evidence and act as an aid to the understanding of the evi-dence. Most, therefore, insist that their interpretations are quite indepen-dent of any dominant self-serving theory or master narrative, a judge-ment that accounts for the widespread popularity among historianstoday of the E.H. Carr approach to history. Almost universally amongpractical realist historians – that majority of practising historiansexisting between the two extremes – it is assumed that the function

History as reconstruction/construction 49

of the historian is not only to establish the veracity and accuracy of theevidence, but also to bring all the known and available evidence into aninterpretative fine focus by employing some organising concepts. Thedissenting left, for example, use class, race and gender in a variety ofways. At the most sophisticated level gender, for illustration, is viewedas a richly complex category of analysis that is best employed when itrecognises the equally important shaping power of the other categoriesof experience. The ultimate object for all practical realist historians, nomatter how complex their methods or whether they ideologically dis-sent or conform, is to use evidence to demonstrate that the conceptsthey use are intrinsic to the evidence.

This situation e!ectively blurs any sharp division between recon-structionism and constructionism. It means in practice that historiansdo not go about their task in two separate phases of research in thesources for the facts, and then exercise interpretation using various con-cepts or models of explanation. Rather the historian gets going, inCarr’s own words, ‘on a few of what I take to be the capital sources’ andthen inevitably gets the itch to write, by which he means to composean interpretation, and ‘thereafter, reading and writing go on simul-taneously’.28 For Carr this means that the feared dichotomy ‘of anuntenable theory of history as an objective compilation of facts . . . andan equally untenable theory of history as the subjective product of themind of the historian’ is far less of a problem than conservative recon-structionists – either of the left or right – might fear. For Carr this ishow people operate in everyday life, a ‘reflection of the nature of man’as he suggests.29

This merging of the mainstreams around the E.H. Carr position isfound in the ideologically liberal practical realism of Appleby, Huntand Jacob as they conclude: ‘Inferential evidence of invisible structuresand patterns abounds’ in history writing today.30 Not surprisingly, theMarxist Alex Callinicos agrees, believing that historical facts are arrivedat ‘inferentially by a process of interpreting data according to a compli-cated system of rules and assumptions.’31 Where the two mainstreamsbranch is often only in the ideologically informed nature of the explana-tory frameworks chosen to interpret the facts. When Appleby, Hunt andJacob argue for the importance of ‘structures and patterns’, they areasking the perennial question: what is the nature of the relationshipbetween free will and determinism in explaining the past? Answers tothis question tend to hinge on ideological preferences. They suggest thatthe forces, the social structures and patterns that influence our lives arerarely palpable, and never quite as simplistic or reductive as, say, a crudeMarxist class analysis suggests. As they argue, ‘The falling rain is

50 History as reconstruction/construction

visible, but it takes meteorologists to explain the structure of climaticchange.’32 Social structures are taken to refer to consistent patterns thatcan be found in behaviour and beliefs and which to a greater or lesserextent determine intentional social action. Without concepts and cat-egories like class, gender, race, nation, city, and so forth, the complex-ities of the past would be inexplicable, remaining at the level of lists ofevents and time charts.

That it is at the ideological level that members of the mainstreamdiverge is evidenced in Callinicos’ refusal to accept the position of theAmerican liberal-relativist and pragmatic philosopher Richard Rortythat historical meaning is at best provisional because, Rorty argues,there is no inherent truth to be discovered in the evidence. Callinicosrejects the Rortyan position that preference between explanatory the-ories may be purely aesthetic. In this way Callinicos rejects the pragma-tic ideologically liberal reconstructionism of Appleby, Hunt and Jacobwhich accepts that historical interpretation may be measured not byreference to a reality found in the evidence and engineered by socialtheory, but according to other ideological criteria with which he doesnot agree. For Marxists in general the truth is indeed ‘out there’, and itis Marxist rather than bogus bourgeois liberal. But they would agree, asempiricists, that it is the material nature of the real world that makes theclaims of historians true or false, not the nature of language or repre-sentation. Correspondence theory for mainstreamers, regardless ofideology, defines what occurs in the real world when our statements, asCallinicos says, ‘capture the way the world is’.33

The constructionist process assumes – regardless of ideological pref-erence – that explanatory frameworks as suggested by the evidencemust be stated in propositional terms that are then open to verificationthrough the further study of the evidence. To take the constructedsocial category of class by way of illustration, historical explanationsusing some kind of class model are legion, and historians create evermore complex theories of class explanation for their own personal use.They normally borrow existing models from colleagues (in history, econ-omics, sociology, anthropology and cultural theory) and then look tothe evidence to refine them as their preferred explanations. As noted atthe start of this section, constructionism is a loose description thatcovers a spectrum of impositionalist approaches to the past. The pre-cise nature of the class model employed by any single historian is, there-fore, dictated by the complexity and assumed explanatory power of thesocial science and cultural models of class-based human behaviour thathe/she has picked over. Other historians inclining towards the recon-structionist mainstream will stick to the empiricist method in which

History as reconstruction/construction 51

they have been schooled, leaving the propositional theoretical dimen-sion at the mundane level of working explanations (but still ready to bemodified as the evidence dictates), rather than go in for very complexsocial science constructionism.

The plea for complex model-making in history was made by JamesHarvey Robinson in his appropriately titled book The New Historypublished in 1912, in which he argued for the study of a much broadersocial history, rejecting the then predominant distinction between his-tory as a method concerned to explain unique events, and other discip-lines that sought general explanations.34 Robinson, however, went onlyso far, fearing to make history the ‘prisoner’ of a priori hypotheses thatmight deny the historian’s objectivity.35 For Robinson and his col-leagues among the French Annales school, empiricism remained thefoundation of the constructionist enterprise, although they acknowl-edged the complexities in the relationship between knower and known,explanation and event. For conservative reconstructionists, the NewHistory marks the start of the descent into relativism. Most historiansin the twentieth century have generally refused the siren voice of grandtheory or deductive positivism, preferring instead to concentrate onthe collection of detailed evidence upon which they could deploy theempiricist inductive method and their low-grade concepts or categoriesof analysis like class. However, the Annales school in France developedthe constructionist tradition of marrying inductive inference from fac-tual evidence with deduction (deductive inference) based upon moregeneral prior sociological generalisations about the socio-economicand politico-cultural structures of society. For its adherents, thisdevelopment added greatly to the explanatory power of history.

Although the point at which reconstructionism historically evolvedinto constructionism is not easy to identify, in the founding of thejournal Annales in 1929 we have as distinct a point of mutation as any.For the first time in the twentieth century, history was written from anexplicit propositional social theory position. From the early seven-teenth century and the advent of the Enlightenment, when reason,experience and science became supreme, generations of European his-torians built the discipline upon the search for truth. Science, likenature, is neutral, rational, factual, logical, unemotional, value-free,calculable and, above all, secular – innocent of man’s faith, religiousdogma or corruption. Although Annales history was designed to be likethat, early practitioners Febvre and Bloch recognised that it could neverbe based on first-hand experience, observation or experiment, sincethere was no Cartesian calculus or geometry in historical knowl-edge. So, while science continued to rely upon empiricism to refine its

52 History as reconstruction/construction

hypotheses (as it still does), history, if it widened its explanatory theor-ies, could also now depend upon other more rigorous mathematical,experimental and observational techniques for the confirmation ofdeductive knowledge.

Since the founding of the school, Annaliste historians like FernandBraudel, Emmanuel Le Roy Ladurie and, more recently, Roger Chartierhave all employed highly sophisticated theories of varying kinds –sociological, economic, cultural, anthropological, psychological andlinguistic.36 The analysis of the ‘structures’ underlying the surface phe-nomena of history, and of the relationship between human intention,agency and action, has not, of course, been limited to the Annalesschool. It has constituted what the historiographer Christopher Lloydhas called a ‘broad structurist tradition of structural history writing . . .which is far from being a school of single coherent approach’.37 Con-structionist history as a result is characterised today by its often greatcomplexity and sophistication, but also by its clear rejection of whatPhilippe Carrard describes as ‘event history’ or interpretations thataccount only for dramatic single and unrepeatable events.38

The high tide of constructionist empiricism arrived in 1942 with CarlHempel’s positivist-inspired article on ‘The Function of General Lawsin History’ in which he claimed that to explain any historical event thehistorian should subsume it under a general or covering law.39 Coveringlaw theory maintains that the historical event ought to be capable ofprediction given the specification of certain contextual conditions. His-tory, like science, therefore, can operate general or covering lawswhich work according to the deduction of the meaning of the event (theexplanandum) from statements consisting of the general law and ante-cedent conditions (the explanans). Hempel recognised, however, thatbecause historians do not really work in this precise manner of formu-lating and articulating general laws, what they do in e!ect is produce‘explanation sketches’ that require ‘filling out’ or refining until theoperative laws of human behaviour become distinct.40 It is only by thisstrict deductive process that history can claim to reconstruct the past.Reconstructionist historians, and their philosophical supporters likeMcCullagh, have nevertheless consistently rejected Hempelian coveringlaw theory, viewing it as deterministic and as an unnecessary distractionfrom their empirical research into the sources and the derivation ofunique historical facts. General laws, if they operate at all, do so formost historians at the far less rigorous level of assumptions which maycover only one instance or unique event. Equally, it is argued that anysingle event is likely to be the result of many so-called laws. Most his-torians today would not rely, as did Frederick Jackson Turner, for example,

History as reconstruction/construction 53

on a single theory to explain the significance of the frontier movementin American history – in his case the existence of free land in the West.When writing history, the historian has to take into account not onlythe contingent but also the di"culties of accurately deciphering humanintentions.

Although people do not always act rationally, the complexities of thepast still continue to be studied by constructionists employing evermore elaborate models of social and cultural institutions, which try totake into account ecological changes, gender redefinitions, classrelations, race, colonisation and de-colonisation, industrialism andtechnology. All require more tools of analysis than simple inductiveinference alone provides. The list of thinkers influencing constructionisthistory now includes the sociologist Anthony Giddens, with his theorythat the historical agent and social institutions are produced by com-plex hierarchies or levels of social practices; the Weberian sociologistsErnest Gellner, Charles Tilly and Cli!ord Geertz, who have applied theideas of social anthropology to historical change; the ecological per-spective of the historian W.G. Hoskins; the total history of theAnnalistes Fernand Braudel, Emmanuel Le Roy Ladurie, Robert Darn-ton and Roger Chartier; the recognition of power structures in societyby the American Marxist social historians Harry Braverman, HerbertGutman, David Montgomery, James Weinstein and Gabriel Kolko; orthe Gramsci-influenced Marxist history of Eric Hobsbawm and EugeneGenovese, as well as that of the Marxist-feminist perspectives of SheilaRowbotham and Catherine Hall.41 These are just a few examples of thehuge range of constructionist explanations available today that seek tolocate the structures influencing apparently unique events.

All these historians grouping around the reconstructionist–constructionist axis still insist on the interrogation of sources to explainhow events happened as they did. Hard-core empiricist opposition tothe analysis of structures over events has been successfully confrontedthrough the insistence that history as empiricism and social theory asproposition cannot be separate in practice. Something else that linksmany among these historians is the fact that while they accept languageto be the vehicle for the ‘concepts’ and/or social theories deployed, mostagree that the explicit and scrupulous defining of the terms, conceptsand categories regularly employed will usually overcome any lurkingproblem of significatory collapse, and most refuse to entertain thenotion that their emplotments may have a major e!ect on the nature ofthe past they purport to discover. As Elton notes, the assumption by the‘theory-mongers’ that language is a dangerous terrain full of pitfalls forthe unwary is not new, and every historian worth his/her salt has known

54 History as reconstruction/construction

about them, and has for many years talked about them – but in a jar-gon-free language we can all understand! So, what is the role of narrativein mainstream history beyond it being merely the medium of report?

HISTORY AS NARRATIVE

The function of language in creating historical understanding centreson the nature and use of narrative. While most historians agree thathistory is in large part a literary process, they disagree over the char-acter and implications of its literariness, but specifically over the ques-tion of whether or not its literary form creates the past as we write it.That history is the discovered truth about the past has been the mainbelief underpinning the general rejection, especially by conservativereconstructionists, of social theory constructionist history, but it hasprovided the rationale for not regarding narrative in and of itself as aform of explanation and understanding. Constructionists tend to viewnarrative as non-scientific and non-explanatory because of its teleo-logical nature, that is, as an explanation orientated towards an alreadylikely known, if not blatantly desired, end result. Constructionists alsoperceive narrative history as inevitably emphasising the unique eventover the discovery and recognition of patterns, because of the stress onthe role of individual people in the past rather than on groups’behaviour or processes.

The rediscovery of narrative has, however, been a feature of recentdevelopments in history-writing. Consequently, some historiansincreasingly view narrative in historical understanding as explanatoryin as much as it is mimetic. As the American historian J.H. Hexter hassaid, narrative displays history’s ‘capacity to convey knowledge of thepast as it actually was’. Most importantly for constructionist historians,Hexter insists that narrative does not deny objectivity because historicalinvestigation, when properly conducted, can produce close approxima-tions to the truth precisely through the discovery of patterns in theevents of the past. As he claims, trying also to satisfy reconstructionists,‘reconstructing connections among the records’,42 suggests that historyis a narrative reconstruction of the past that can objectively reveal whatactually happened. He concludes:

The function of the historian’s language . . . may best be described as‘translational’; it aims to assist the reader to translate his experiencefrom a familiar accepted context into a context strange and perhapsinitially repugnant. The direction of the translation is as importantas its e!ectiveness.43

History as reconstruction/construction 55

This recognition of the way in which historians use narrative to director ‘translate’ meaning is unproblematic for Hexter because it is an inte-gral part of the constitution of historical interpretation. This historianis the guide and narrative the terrain.

M.C. Lemon agrees that historians communicate primarily throughlanguage’s written narrative form; as he says, they ‘trans-form’ theirthoughts into language. Translating thinking into language, however,does not prove the correspondence theory. What it does do is reinforceand support the basic historical method of inductive inference. AsLemon says, the reader has to infer ‘from what is said, the thinking itevidences. This is the function of narrative’.44 In order to establish thethought behind the primary or secondary evidence, readers and his-torians must first understand the language used. This logic follows theCollingwood–Carr approach noted already. The main di"culty with itis the question of the extent to which language is constitutive of realityrather than its reflection. For practical realist mainstreamers this is nota major problem because they assume that narrative is not the primarymechanism for historical explanation – historical explanation emergesinferentially from the study of the sources and/or the use of analyticalmodels of explanation, not from ‘this happened, then that’. Althoughnarrative conforms to this basic structure of change over time, this isprobably not a good enough basis on which to claim that it is theessence of historical explanation. Narrative is taken to be the form inwhich historical analysis is transmitted to its readers, but the furtherclaim can be disputed. While narratives may carry or contain explan-ations, they are not explanations in themselves.45 The issue remainswhether or not the historian thinks language simply reflects reality or isthe major element in constituting how we understand it.

Since the 1970s, however, the choices that historians make about theirdescriptions, emplotments, figurative styles, the construction of theirexplanatory arguments and any moral judgements with which theyengage have been increasingly discussed and acknowledged as signifi-cant features of narrative explanation. The practical realist mainstreamposition is clearly stated by Lawrence Stone in his 1979 article. Afterdefining narrative simply as ‘the organisation of material in a chrono-logically sequential order and the focusing of the content into a singlecoherent story’, he says that narrative history di!ers from social theoryor constructionist history (what he called ‘structural history’) in that ‘itsarrangement is descriptive rather than analytical and that its centralfocus is on man not circumstances. It therefore deals with the particularand specific rather than the collective and statistical.’ For Stone,economic determinism, structuralism, quantification and psychohistory

56 History as reconstruction/construction

are poor substitutes for a narrated empiricism that produced historicalunderstanding ‘based on observation, experience, judgement andintuition’.46

Although the object of Stone’s defence of narrative was to attack‘the attempt to produce a coherent scientific explanation of change inthe past’,47 as we noted in Chapter 1, in the mid-1960s William Galliehad argued for the centrality of narrative as the characteristic form ofhistorical understanding, and, like Hexter, had suggested that narrativeand constructionism are not incompatible. Historians understand thepast as they produce a story that they and their readers can follow,based on the causally related and contextualised evidence. Gallieargued that to follow a historical narrative may regularly require theacceptance of explanations that stretch one’s credulity.48 What he issaying is that no matter how unlikely is the story of a series of eventsand their changing relationship over time, if it is reasonably supportedby the causally connected evidence, it should be believed. M.C. Lemon,however, insists that no matter how unlikely it may appear, narrativehistory is not about establishing constructionist-type mechanisticcausal relationships between events. Its explanatory power arises fromits in-built ability or power to track, trail or follow the individual per-son’s intentional responses to their context. The function of narrativehistory lies in discovering the intentionality of people in the past andduplicating it through narrative, making the story followable andcomprehensible.

More recently, the French historian Philippe Carrard has commentedon how the highly sophisticated New History continues to rely uponnarrative as its primary vehicle for expression and explanation. Heclaims that ‘New Historians . . . still depend largely on storytelling tomake sense of the world . . . [and] . . . this analytical component is stillframed by a plot, and this plot retains essential cognitive functions’.49

Like William Gallie and Arthur Danto, Carrard is insisting here thateven constructionist history requires emplotments that identify theintentions of the historical agent and which also constitute explan-ations. Writing history seems to be a highly complex amalgam of inter-pretation of the sources (discovering how people in the past acted withintentionality), plus overt propositional testing (using abstract socialconstructions like class), translated into an understandable andexplanatory narrative.

As we shall see further below, the work of the Dutch historical the-orist Frank R. Ankersmit is central to the debates on the narrative char-acter of history. His rejection of reconstructionist foundationalism’sidea that history is only ever determined by ‘what happened’ has made

History as reconstruction/construction 57

an intellectual space for a far more sophisticated understanding ofhistory as a literary meaning-creation activity. As Ankersmit insists,understanding how description and representation work is crucial tograsping how history works. It is a common oversimplification, one thatreconstructionists are, by definition, always subject to, to suggest thatmeaning and truth in history derive only from the raw data of events.This is not to say the real is being reduced to text. Rather, it is suggestedthat the meaning of ‘the reality of the past’ can only be apprehendedthrough the texts we create about it. As Frank R. Ankersmit reminds us,the historical narrative is a substitution for past reality and one that, inits creation, is not insulated from either our present or the ontologicalnature of its construction.50 Historical narratives are, strictly speaking,proposals about the past with language substituting for reality. It can-not be otherwise. This being the case, as Ankersmit suggests, it demandshistorians to take this into account when thinking about the cognitivestructure and power of historical narratives. While reconstructionistsremain unreconstructed in their rejection of this notion, the majority ofhistorians have begun increasingly to acknowledge the narrative logicof history.

Joyce Appleby, Lynn Hunt and Margaret Jacob support the main-stream consensus by arguing in favour of the necessity for historians tocombine ‘narrative coherence, causal analysis, and social contextualisa-tion’, creating a process that they believe is ‘exemplified in our ownnarratives’.51 Mindful of their empiricist credentials, they reject ‘thecurrent negative or ironic judgements about history’s role’ of the post-modern deconstructionists while carefully recognising the aesthetic orliterary choices that historians have to make when writing history. Theysummarise the judgement of the mainstream on the literariness of his-tory when they judge its literary dimension as not being history’s firstconsideration. Their ranking of primary choices as narrative historiansare ‘political, social and epistemological’, and reflect their beliefs abouttheir role in the community of historians and the nature of Americansociety. While agreeing ‘that the focus on culture and language under-mines this hierarchical view by showing that all social reality is cultur-ally constructed and discursively construed in the first instance’, theystill maintain their belief in the practical knowability of past reality. Inaccepting that narrative ‘is a universal mode of organising humanknowledge’ and that there is a gap between ‘reality and its narration’,narrative remains, nevertheless, ‘an inappropriate vehicle for historicalexplanation.52

Taking narrative to be a form of report rather than of knowing is ajudgement reached by philosopher of history Michael Stanford. It is his

58 History as reconstruction/construction

view that ‘history does not have to be narrative history.’53 As he pointsout, the commonsense view that life happens like a story is merely aploy of writers and nothing more. Events do not, in reality, conveni-ently occur in a narrative form. This is supplied by the historian later,but what is really important, he says, is to note that ‘most academicworks of history are not written in the narrative mode’.54 The reasonStanford gives for this in opposition to Gallie is that the narrative formcannot cope with the complexities of causally related events. The inter-weaving of the political, cultural, social and economic is so complicatedthat description alone cannot substitute for constructionist-typeconceptual analysis.

As the result of the popularity of this view, most history writtentoday is done through a topics or problems approach rather thandescribing single events in a sequence that the historian assumes will, ine!ect, become self-explanatory.55 In examples taken from my o"ceshelf, Phyllis Deane, in her undergraduate British economic historyprimer The First Industrial Revolution (1965), reconstructs Britishindustrialisation through topics such as the Demographic Revolution,the Transport Revolution, the Iron Industry, the Role of the Banks, andStandards of Living. A second example I found from European econ-omic history is Clive Trebilcock’s The Industrialisation of the Contin-ental Powers 1780–1914 (1981), which is constructed around modelsof industrialisation associated with individual European countries.Another example taken at random, but this time from recently pub-lished American history, is Vicki L. Ruiz and Ellen Carol DuBois’Unequal Sisters (2000), a multi-cultural reader in American women’shistory from the colonial slave days up to the 1990s, which is also organ-ised around distinct topics – housework and ante-bellum working-classsubsistence, the marriage structure of Mission-educated Chinese-American women, the cosmetics industry and the construction ofgender, and Vietnamese immigrant women.56 The sequencing ofevents over time is handled di!erently in all three of these texts – whileit is there it is always of secondary importance to constructionistexplanatory frameworks.

So, although its nature remains disputed, the recovery of narrativehistory by the mainstream is a well-established trend. Conservativereconstructionists are willing to defend narrative only as the vehicle forconclusions inferred from the sources. Practical realists, probably alongwith the majority of constructionists, maintain that narrative carriesmeaning but remains secondary to their conceptualisations andexplanatory social theories. None, however, accept narrative as either awholly unproblematic descriptive vehicle, or so fluid in meaning as to be

History as reconstruction/construction 59

incapable of conveying any definitive knowledge. Reconstructionistsand constructionists see no good reason to believe that just becausenarrative is not the primary instrument for creating historical knowl-edge it is a worthless mechanism for conveying the results of historicalresearch.

CONCLUSION

What I have argued in this chapter is that the reconstructionist/constructionist mainstream approach depends on several related prin-ciples. The first is the acceptance of an objectivist-inspired method-ology, which in serving the evidence and isolating the historian allowsfor the accurate, independent and truthful reconstruction of the past.Second, it follows that the truth of history can be distinguished fromfiction and value-judgement, with history being about discovering whathas actually happened. I noted, however, the divisions within the re-constructionist/constructionist mainstream with Elton’s attacks on allforms of history produced by the fashionable ‘theory-mongers’.57 I alsostressed the argument of McCullagh in support of Elton, that it ispossible to believe ‘in the truth of historical descriptions’.58 In contrastto Elton I pointed out the Callinicos position, that facts emerge fromtheory-inspired historical study, and how recently the debate hasfocused on whether historical narrative can be considered in and ofitself to be a form of explanation.

We should now be in a better position to understand the four mainpremises of the traditional or reconstructionist school: that historypossesses its own epistemology; that historical method consists of theforensic examination of primary sources according to the inferentialrules of evidence (comparison, colligation, verification and theimpartial interpretation of that evidence); the rejection of general lawsin as much as they imply history can be predictive; and finally, thatnarrative as the medium for historical reconstruction, although it is notan adequate form of explanation, is not an obstacle to the enterprise. Inthe next two chapters I shall appraise these mainstream premises fromthe perspective of the deconstructive consciousness.

60 History as reconstruction/construction

4 History as deconstruction

INTRODUCTION

The history profession is not starkly divided between deconstruction-ists and the reconstructionist/constructionist mainstreams, not leastbecause, as we have seen, there are active debates that cut across allpositions, and most historians presuppose the use of narrative at leastas the vehicle for conveying historical knowledge if not for creating it.But it is on this very point that there is still a broad division betweenthose who think self-consciously about the nature and particular roleof narrative in the practice of the craft, what I have designated as thedeconstructive consciousness, and those who view the reconstructionof the past as primarily a skilled engagement with the evidence andwho think, therefore, that there is little to dispute about its writtenform as history. As I have indicated, this division focuses on howcontent and form relate, specifically the extent to which historicalknowledge and explanation are the function primarily of evidenceplaced in context or the aesthetics and structure of narrativediscourse.

Conservative reconstructionist historians do not accept empiricismas only one of several competing modes of knowing the past. Theyreject all other methods of historical interpretation, especially thosethat smack of an ideology of which they disapprove, e.g. Marxism,cultural materialism, Hegelianism, bourgeois liberalism, or whatever.Historians in the mainstreams prefer to view history as primarily apractice – the craft of history.1 It is perceived as a technique of non-ideological discovery.2 What is challenged by the deconstructive histori-cal consciousness is this belief that historical investigation can o!er apeculiarly empiricist historical litmus test of knowledge, emphasisinginstead the belief that the past is only ever accessible to us as atextual representation – ‘the past’ translated into ‘history’. From a

deconstructive perspective on the significance of language andnarrative structure, I will now address each of the four questionsin turn.

EPISTEMOLOGY

As a consequence of the post-structuralist challenge to empiricism andthe correspondence theory of meaning we are confronted by what, atfirst blush, appears to be the discomforting notion that there is noaccess to knowledge except through the murky and dangerous waters oflanguage. Historians as a group respond to this by refusing to exploreits implications. In spite of Derrida’s and Barthes’ warnings, historiansgenerally continue to rely on the commonsense notion that they willlocate the knowable external presence to the text in the context. This isthe investment the discipline has in referentiality – a referent for eachword and consequently a precise meaning to be discovered. The prob-lem is that such a fixation makes it very di"cult to view narratives forwhat they are: meaningful historical explanations in themselves, ratherthan plain vehicles with which to explain the past as it actuallyhappened. In order to pursue this we need to know more about hownarrative works in epistemological terms.

This opening up of historical analysis to questions of rhetoric isfound in the work of Hayden White and other philosophers and his-torians like F.R. Ankersmit, Hans Kellner, Jörn Rüsen and Keith Jen-kins. The deconstructive historical consciousness suggests that historywritten by working historians should explicitly acknowledge and, whenappropriate, explore its emplotted or prefigured form. What is arguedfor is that the analysis of style, genre and narrative structure, moreusually associated with fictional literature, be applied to the under-standing of the historian’s sources and written interpretations.Although this approach emerges from structuralism’s early concernwith the arbitrary nature of language, history produced within thedeconstructive consciousness has a much wider range of concerns.Reconstructionist historians choose, however, to keep both structural-ism and historical deconstruction at arm’s length by regarding the writ-ten form of the past as somehow not especially relevant to thereconstruction and explanation of the past as it actually was. Althoughthey applaud precision in the use of language and recognise its limita-tions, the importance of language-use in its broadest explanatory senseremains secondary to the discovery of true origins, causal analysis andcontextualism.

As I have already indicated, the early seventeenth-century positivist

62 History as deconstruction

legacy of Francis Bacon has remained the controlling metaphor oftwentieth-century historical study even at the practical realist centre.History becomes genuinely problematic only when historians drawuntenable inductive inferences, shape history for their ideological/political purposes or, what for a few is worse, dabble in the nether worldof hypothesis-making. History should be like science to the extent thatscience is the study of the real world ‘out there’, is factual not specu-lative, empirical rather than a priori, verifiable, anti-hypothetical,ideologically neutral and, above all, non-impositionalist and objective.Consequently, the fundamental implication of the theories of post-modernism for history – its demise as a legitimate discipline – isunacceptable.

Questioning history as an empirical project ought not, in fact, be aproblem for historians. If we accept that there are no master narratives– such as history proper assumes itself to be – then, as Lyotard says,there is no inside track to reality. Questioning the epistemological basisof history, however, cuts deep into the mind of historians. It concernsthe objectivity with which the historian deals with sources and thenwrites up a disinterested interpretation tracing and explaining originsand causes. While most historians would not argue that historicalmethod is scientific, there remains this strong sense of being rationallyand objectively in touch with a potentially understandable, causallyanalysable and truthful past.3 To argue otherwise is simply to cease tobe a historian.

The leading critic of what we might loosely call traditional historyis Michel Foucault. In accepting the German philosopher FriedrichNietzsche’s reaction against the certainty of the empiricism of the sec-ond half of the nineteenth century, Foucault’s attack on history is lessdirected towards the post-structuralist indeterminacy of language, butrather more against the manner in which historians believe in therecovery of the truth of the past.4 Foucault challenges the belief thathistorians can e!ectively step outside history, capture the context, andbe objective – arguing instead that all written history is an act of cre-ation through the narrative impositionalism of the historian as he/sheemplots the data, and this act is to some degree the ideological productof the age in which he/she lives.

Foucault’s critique of history as a legitimate discipline is paralleledby the French cultural critic Roland Barthes. Building on the distinc-tion between histoire, in which events seemingly tell themselves withoutthe intrusion of a narrator, and discourse, which is overtly self-conscious and authorial, in his essay ‘The Discourse of History’ (1967)Barthes contests history’s dependence on the correspondence between

History as deconstruction 63

evidence, the designation of historical facts and the ‘reality e!ect’ of‘objective’ history as created in the historian’s written interpretation.5

Barthes suggests that written history is only another narrative,e!ectively collapsing the story discourse distinction.6 As Barthes’ inter-preter Stephen Bann comments, the ‘rhetorical analysis of historicalnarrative . . . cannot grant to history, a priori, the mythic status whichdi!erentiates it from fiction’.7

In his defence of narrative in ‘The Discourse’, Barthes strikes at thevery existence of history as an epistemology. History, he notes, is usu-ally ‘justified by the principles of “rational” exposition’ but he asks‘does this form of narration really di!er, in some specific trait, in someindubitably distinctive feature, from imaginary narration, as we find itin the epic, the novel, and the drama?’.8 He goes on to challenge theauthority of the historian based on his/her access to the sources, byemphasising that the real work of history resides in their translation(Barthes describes this as utterance) into a narrative of historical inter-pretation. Barthes’ challenge takes the shape of a critique of the struc-ture of the historian’s discourse. The examples he o!ers include thehistorian’s traditional deployment of lots of detail amid the minutiaeof events. In the history of art this is the trompe l’æil principle wherebyfine detail is intended to create a sense of reality. Barthes’ challenge alsoextends to how historians complicate chronology by compressing timein a few pages, flipping back and forward through the past. Barthes’further probes the historian’s unspoken claim to omniscience – the pro-cess whereby the historian absents him/herself from the discourse tocreate the impression of realism through direct access to the referent –from where, as Barthes says,

there is in consequence a systematic deficiency of any form of signreferring to the sender of the historical message. The history seemsto be telling itself all on its own. This feature . . . corresponds ine!ect to the type of historical discourse labelled as ‘objective’ (inwhich the historian never intervenes). . . . On the level of discourse,objectivity – or the deficiency of signs of the utterer – thus appearsas a particular form of imaginary projection, the product of whatmight be called the referential illusion, since in this case the historianis claiming to allow the referent to speak all on its own.9

The epistemological status of historical discourse is thus conventionallya"rmed and asserted. The historical fact is privileged by being placedin the specially reserved position of a superior claim to truthfulness, aswarranted by both a plain language and an independent research meth-odology and as supported in the notes and references – the sca!olding

64 History as deconstruction

of proper historical methodology. Barthes goes on to suggest that thisillusory correspondence between plain language, historical evidenceand historical truth is also to be found in realist novels which similarlyappear objective because they too have suppressed the signs of the ‘I’ intheir narrative.

Barthes is claiming that historians play a confidence trick because ofthe way in which we use the trope of the real – in e!ect the Eltonmethod – to wring historians out of history and presume to get to thereality of the past. Barthes is suggesting that history is performing anepistemological trick through which the referent is placed in a privil-eged world of the real beyond arbitrary signification. As he says, ‘Thehistorian is not so much a collector of facts as a collector and relator ofsignifiers; that is to say, he organises them with the purpose of establish-ing positive meaning.’10 While most mainstream historians accept anorganising role for the historian, they draw the line at this decon-structive vision which holds that there can be no objectivity in selectionof material, and that all judgements about what to include or excludeare based on ideology, preferred narrative structures, and the limita-tions of the signifier–signified–sign relationship. Barthes’ decon-structionist point is that the historian deliberately confuses or conflatesthe signified with the referent, producing a signifier–referent cor-respondence, hence Barthes’ warning that ‘in “objective history” the“real” is never more than an unformulated signified, sheltering behindthe apparently all-powerful referent. This situation characterises whatwe might call the realistic e!ect ’.11 This is similar to Foucault’s idea thatall discourses are at best perspectives that produce truth e!ects. This isnot so much anti-referentialism as a recognition of referentialism’sboundaries.

Most historians refuse to view the real as only a truth-e!ect, giventhe profession’s continuing investment in the independence of the dis-cipline and the traditional Western belief in reason and rationality(logocentrism). In so doing we fail to acknowledge that the narrativedescription of historical facts is integral to our proof of those facts.Barthes comments that by instituting ‘narration as the privileged signi-fier of the real’, historical truth emerges as the composite of ‘carefulattention to narration’, and the ‘abundance of . . . “concrete details” ’.He concludes that ‘Narrative structure, which was originally developedwithin the cauldron of fiction (in myths and the first epics) becomes atonce the sign and the proof of reality’.12 These are concerns that influ-enced Hayden White, among others, to explore the rhetorical dimen-sion to writing history, and have posed a question mark over narrativestructure and the impositions it makes on writing history.13 Regardless

History as deconstruction 65

of Barthes’ argument that history is at best a fudged-up performativeand unavoidably ideological, mainstream historians still insist that theywork in a discipline that aspires to a high degree of correspondencewith the past as it actually was and that narrative is a vehicle for reportrather than the primary (if flawed) medium of explanation. Decon-structionist historians, however, are driven to ask what kind of epi-stemological status can the sorts of stories historians tell have, and whathave they the right to claim, by virtue of their narrative form?14

EVIDENCE

There are two related questions raised by deconstructionist historyabout historical evidence. How can we discover the intentionality in themind behind the source, and how much reliance can we place on thereconstructionist’s contextualisation of events as a form of explan-ation? It is here that we come across the apparently strange notion ofthe death of the author/subject. For Barthes, the importance of theauthor of historical evidence is diminished in as much as he/she isperceived as representative of further texts and ideological positionsrather than as the originator of meaning. Evidence does not refer to arecoverable and accurately knowable past reality but represents chainsof interpretations, that is, we have no master or transcendent signifiers.In the sense that we as historians cannot know what were the intentionsof the author of the source, to suggest that we look to those intentionsas a means to interpret the evidence is only to invite yet further textualinvestigation. This contradicts Lemon’s view that narrative’s power toexplain emerges from its tracking of the historical agent’s intentionaland intelligible response to their context. Barthes maintains that

The names of authors or of doctrines have here no substantial value.They indicate neither identities nor causes. It would be frivolous tothink that ‘Descartes,’ ‘Leibniz,’ ‘Rousseau,’ ‘Hegel,’ etc., are namesof authors, of the authors of movements or displacements that wethus designate. The indicative value that I attribute to them is firstthe name of a problem.15

The inevitable rejection by empiricists of this position is founded on thebelief that the historian and the evidence are separate entities – a fur-ther re-stating of the traditional distinction between knower andknown – and this gap permits historians to stand back and see theorigins of meaning in the evidence.

F.R. Ankersmit describes what he calls the postmodernist historian’sperception of evidence as a tile, not to be picked up to see what is

66 History as deconstruction

underneath it, but as something which the historian steps on in order tomove on to other tiles: horizontally instead of vertically.16 For HaydenWhite, this perspective (stepping from tile to tile) has further signifi-cance for the constitution of meaning because of what it says aboutideology.17 The real problem with historical evidence for White isnot Barthes’ unending roundabout of meanings, but the inevitableideological dimension to the interpretation of evidence.

The idea of historical interpretation being influenced by ideologicalconsiderations seems wrong to reconstructionist historians. Elton, forexample, rejects any ideological impositionalism of the historian of thekind acknowledged by White because it produces ‘uncertainty aroundhistorical truth’. The ‘true view of the past’ emerges instead for Eltonfrom ‘the deficiencies of the evidence and the problems it poses, ratherthan from the alleged transformation of events in the organising mindof the historian’.18 White opposes this, persevering with the argumentthat

there is no such thing as a single correct view of any object understudy but . . . there are many correct views, each requiring its ownstyle of representation. For we should recognise that what constitutesthe facts themselves is the problem that the historian, like the artist,has tried to solve in the choice of metaphor by which he orders hisworld, past, present, and future.19

Any crossing of the boundary between the observer and that which is tobe observed, through the choice of metaphor, thus clearly contravenesone of the most basic ‘rules’ of traditional historical analysis because itthreatens Elton’s ideal of objectively dealing with the evidence. Becauseobjectivity is the central metaphor of empiricism, the ideological mesh-ing of historian and his/her sources starkly presents the danger of sub-jectivity and eventual corruption of history. Even R.G. Collingwood’sinterventionist historical method, ‘that the historian must re-enact thepast in his own mind’, presupposes a minimum level of objectivity. Ihave already pointed out how this also prompts the argument that, by athorough knowledge of the facts, the reconstructionists reject the follyof social science model-making as applied to history, notably the use ofsocial theory and the appeal to covering laws.20 While the questionof subjectivity in dealing with the evidence is at the heart of the longcontested issue of covering laws in history, it is also a debate that isimportant for the deconstructive consciousness. It further opens upthe epistemological foundation of narrative as a legitimate type ofexplanation in distinction to, among other things, overt socialtheorising.

History as deconstruction 67

THEORIES OF HISTORY: CONSTRUCTING THE PAST

To his own question, ‘Of what can there be historical knowledge?’, Col-lingwood’s reply, ‘Of that which can be re-enacted in the historian’smind’, remains a problem for many reconstructionists because itis not based upon their method of historical analysis.21 Collingwoodelaborated, ‘Of that which is not experience but the mere object ofexperience, there can be no history.’22 To overcome the lack of directexperience in historical explanation Collingwoodian historians likeE.H. Carr plunge themselves in the evidence and experience the pastas best they can – by rethinking it. Although crude empiricists likeGeo!rey Elton believe this to be a quite wrong-headed method –believing instead in maintaining the distinction of knower and known– they would generally agree with Collingwood that, whatever methodis used, historians must avoid the more grievous error of appealing toa universal explanatory social theory that is usually just a fancy cloakfor personal bias or the methodological dead-end of covering lawpositivism. The framing of laws in the form of a proposition suggest-ing why an event occurred in order to yield causal connections istaken not to be history.23 But, as Callinicos suggests from his con-structionist Marxist standpoint, the study of how humans relate totheir contexts necessitates a social theory. For Callinicos, all historymust attempt to discover some pattern in the transformations inhuman society.

As we have already noted, covering law theory is unpopular amongthose who judge it to be founded on a model of historical explanationacquired from science. For others, its unpopularity stems from its rele-gation of the power of narrative to explain the past. Consequently, fewhistorians have employed what Hempel designated in the early 1940s ascovering law theory. Some fifty years before, one of the most influentialpieces of written history – Frederick Jackson Turner’s work on the roleof the frontier in American history – illustrates positivism’s limitedinfluence. While denying the existence of general laws in history, Turnerwas largely alone in using them in practice. By borrowing from thesocial and the natural sciences, Turner became one of the leading his-torians of his generation by inferring the existence of a general law thatapplied directly to American historical experience.24 He argued in hisfamous lecture before the American Historical Association gathered inChicago in 1893 that

The existence of an area of free land, its continuous recession andthe advance of American settlement westward explain Americandevelopment.25

68 History as deconstruction

For Turner, this law of westward movement accounted for Americanhistory. Turner’s approach made him one of the leading social scientistsof his time. However, the reaction against positivism in historicalexplanation emerged in the interwar years, led by two more Americanhistorians, Carl Becker and Charles Beard. Derived again from aNietzschean position, but specifically under the influence of Italian his-torian Benedetto Croce, Beard and Becker challenged any objectivisthistory that saw itself as above the concerns of the present.26 Endorsingthis relativist line, Becker asserted that ‘Historical thinking . . . is asocial instrument, helpful in getting the world’s work more e!ectivelydone.’27 Most historians by today have accepted relativism at least tothe extent that they continue to reject absolutist covering laws, butstill refuse to accept that there may be more fiction in history thanpositivists will admit.28

For deconstructionist historians, the rehearsing of these argumentsfor or against constructionism is a rather meaningless exercise if oneentertains doubts about the truth-value of textual evidence and theinterpretation built upon it. Debating covering law theory is irrelevantif the whole empiricist model of induction and inductive inference isflawed, because facts neither measure nor produce the kind of historicalknowledge that mainstream historians claim. Most mainstream his-torians ignore the implications of this, preferring instead to concentrateon the sources, endorsing Collingwood’s description of the historicalmethod as the objective analysis of sources into their component partsto distinguish which are the more trustworthy. However, Collingwoodalso acknowledged the role of the historian in construing historicalaccounts. As he argued, historians know how to do their own work intheir own way and should no longer run the risk of being misled bytrying to assimilate scientific method into history.29 The almost uni-versal rejection of positivist constructionism rests, however, on thedoubtful belief of most historians that historical explanation is reallyobjective interpretation cast in a narrative form. The deconstructiveargument holds instead that our sources are never transcendent signi-fieds because they have a pre-figured historical status by being alreadyrecounted in chronicles, diaries, legends, memories and interpretations,even before another generation of historians go to work on them again.

The deconstructive critique of empiricist representation and referen-tiality e!ectively asks: does knowledge emerge through social beingand/or language-use? Although as a form of representation, narrativealways fails the correspondence test, it remains of crucial importance inthe reconstruction/construction of the past. It is worth considering atthis point that the e!ort to find out truth in the past may be less about

History as deconstruction 69

the rules of evidence, covering laws and even narrative, but is perhapsabout the will to gain power. For Foucault there is a fundamentalchasm between language and reality. The only reality is found whenlanguage produces a meaning. We use language but language also usesus.30 Consequently, narrative is a discourse, the currency of which ispower. That power may well be used to create a usable past for a nation.Narrative may, therefore, be viewed as a discursive formation that existsin the present and is not a simple and uncomplicated reference to thepast.31 The accretion of historical knowledge – ‘knowing the past’ –often justifies the present, or some preferred version of it, and this is themotivating force that drives the historian as a professional. Accord-ingly, Foucault argues that all historians, because we are attached to aprofession and a discipline, have a vested interest – usually ideologicallyframed – in maintaining the importance of the myth of the objectivesearch for truth, whether reconstructionist or constructionist in orienta-tion. The worst o!enders in Foucault’s eyes are liberal bourgeoisempiricists who believe that they have a control over their ideology thatallows them objective access to the essential past. The point of decon-structionist history is the challenge it throws down to the idea, whichreaches its ultimate expression in hard-core constructionism, especiallyof the statistical variety, that there are essential (true) patterns ‘outthere’ to be discovered in the past.

The deconstructive consciousness assumes that the treatment of theevidence in the historical narrative deals mainly in verisimilitude andcoherence rather than objective explanation. This does not mean thatwe are all extreme relativists. White, for example, rejects extreme scepti-cism about the epistemological value of narrative, in fact putting it atthe centre of what history is really about. ‘As thus envisaged, the“content” of the discourse consists as much of its form as it does ofwhatever information might be extracted from a reading of it.’32 Whiteconcludes that a discourse should be regarded as an ‘apparatus for theproduction of meaning rather than as only a vehicle for the transmis-sion of information about an extrinsic referent’.33 In acknowledging thecognitive importance of narrative, White does not suggest that it canrecover the past as it actually was any more than can positivism. Decon-structionist history’s suspicions about referentiality and representationin the reading of sources, and the writing of history, doubts aboutrecovering the intentions of the author, constructionist theorising, andthe often hidden agenda of power not only mean questioning the claimsof the mainstream, but also attest to the need to address more fully theshortcomings, as well as the potential, of historical narrative as a meansof explanation.

70 History as deconstruction

HISTORY AS NARRATIVE

The impact of the deconstructive consciousness means not only ques-tioning historical interpretation as an objective avenue to the past as itactually was, but also entails exploring the explanatory or story-tellingpower of narrative. This was made clear by Amy J. Elias when she saidfor the past thirty years history has debated itself and for the momentthe new empiricism (she does not use this term precisely) has tried towed two incompatible ideas: empirical truth and post-structuralist lan-guage theory.34 This has generated a pragmatic non-essentialist naturefor history which is probably unsatisfactory all round. If historical writ-ing is the analysis of complex, pre-existing chains of interpretation,whereby documents do not guarantee authorial meaning and signifierscreate only more signifiers, then discussion of content in history muststill begin with an understanding of its linguistic and story form. His-torians are increasingly encouraged to think not only about researchingthe past, but also about how to express and undertake that research.Thinking about the form will make us think about how to deal with itscontent. To what extent then is the form of written history as significantas its factual content?

W.H. Dray summarised the various positions that can be held on theimportance of narrative to historical explanation, namely that

history simply is narrative; or that it is essentially narrative; or thata history must contain some narrative elements; or that one form ofhistory, at any rate, and perhaps the most important one, narrates. Ithas been held, too, that it is through narration that historians achievewhatever is specifically historical about historical understanding; orthat historical explanations get their distinctive structure by reasonof their occurring in the course of historical narratives. It has evenbeen held that narratives can themselves be explanatory in a specialway; or that narrative is per se a form of explanation, if not indeedself-explanatory.35

The functioning of narrative is thus a dilemma for historians. Narrativeclaims to represent the complexities and realities of the past, butbecause it is a story form it must be the creation of the historian’simagination. Can it therefore entertain any claim to being a true repre-sentation of what actually happened? Narrative, Louis Mink suggests,is the product of an ‘imaginative construction which cannot defend itsclaim to truth by any accepted procedure of argument or authentica-tion’.36 This means that historians unavoidably impose themselves onthe past by inventing narratives as they try to explain what the past

History as deconstruction 71

‘really meant’, what the source-text ‘really says’, what the author’sintentions ‘really were’.37

As we know, for most historians narrative is the unquestioned formof history. Although a number of philosophers of history have arguednarrative to be the essential and distinguishing feature of history, mostpractitioners fail to grasp its practical methodological significance, stillregarding it as only a casual stylistic property that some essays possessand others do not. Like most things, whether narrative is explanatory ornot comes down to how we define it. The debate on it as a legitimateform of historical explanation has produced anti-narrativists, amongthem philosophers of history Maurice Mandelbaum and Leon Gold-stein, who claim that although narrative is an element of historicalstudy, not all history has to be framed in the narrative form, and thediscipline has other prior and more important methodological claims.Then there are pro-narrativists like philosophers Frederick A. Olaf-son, David Carr, William Gallie, Arthur Danto and A.R. Louch whoinsist there is a strong correspondence between the past as lived, andhistory as written.38 Then there are those pro-narrative but determinedanti-deconstructionists like J.H. Hexter and Lawrence Stone, who donot accept that language must always fail the correspondence test.Finally, there are those of a broadly defined deconstructive turnlike Hayden White, Dominick LaCapra, F.R. Ankersmit, HansKellner and David Harlan, who view narrative as the essential butlargely misunderstood feature of historical explanation – a misunder-standing that among many other things permits history a claim to aspurious epistemological legitimacy through its favourite metaphor ofobjectivity.

Maurice Mandelbaum, in observing the general relevance of narra-tive, suggests that historians write it while keeping their ‘eyes on higherthings’ – the prize of historical truth.39 Like Arthur Marwick, phil-osopher of history Leon Goldstein cannot understand the fuss madeover history’s narrative form, what he calls the superstructure of his-tory. Its real business is research on archival sources, the infra-structure.For Goldstein, history is ‘a technical discipline’, one that uses methodsthat are peculiarly its own: ‘History is a way of knowing, not a mode ofdiscourse.’40 He concludes, ‘What we know about the historical past weknow only through its constitution in historical research.’41 The decon-structive turn counters this by declaring that the past exists as historyonly because a narrative or story structure has been imposed by thehistorian on the evidence.

Because the historical text consists of a narrative that purports todescribe and evaluate past reality, what is at issue is the power to explain

72 History as deconstruction

of story-form narrative. As we have seen, structuralist and post-structuralist literary theory has thrown open the question of how his-torians employ narrative as a way to fix historical knowledge as uniqueto itself, and consequently divide history from other kinds of writing.42

In support of a pro-narrativist position, M.C. Lemon argues that thelogic of life is replicated in narrative. As he says, the lesson is that ‘thereare, “out there”, amidst a virtual infinity of occurrences real stories tobe truly told and their telling must conform to the logic of narrativeexplanation’.43 Lemon’s view is shared by Dominick LaCapra, HaydenWhite and Paul Ricoeur, who maintain that because of its essentialnarrative form, history cannot be categorised as anything other than akind of literature, but that this does not devalue its significance orexplanatory power. The consequence is, in fact, a recasting of its char-acter and functioning. As Paul Ricoeur says, history must possess an‘irreducibly narrative character’ in the same way that human existencedoes.44 Its function is to describe the process in which people construethemselves and their culture through the production of language. Thisemphasis on the cognitive value of narrative does not of course meanthat we now suddenly have the access to the past as it actually was – weonly have a story version of it. Narrative can explain the past, but notguarantee that its explanations are truthful.

Deconstructionist historians approach this issue through the follow-ing thinking. The past as it actually was, and the individual historicalstatements composing its narrative, can never coincide precisely. Theproblem is that we cannot verify the past by the evidence. Evidence isnot past reality because our access to it must be through many inter-mediaries – absence, gaps and silences, the contrived nature of the arch-ive, signifier–referent collapse, the historian’s bias and, not least, thestructure of the historian’s imposed and contrived narrative argument.It is probably best to view historical narratives as propositions abouthow we might represent a past reality, suggestions of possible cor-respondences rather than the correspondence. Hayden White endorsesphilosopher Arthur Danto’s view that historical facts are really onlyevents under a description.45 It follows that, as events under a descrip-tion, these narrative proposals/suggestions are the result of individualhistorians’ interpretations and compete for acceptance in those terms.History results not from the debate about past reality as such, but fromcompeting narrative proposals about the nature and possible meaningsof past events. Of course, once a narrative proposal has achieved amore or less universal acceptance (like ‘the Cold War’ or ‘the IndustrialRevolution’), it becomes concretised as past reality. It is no longer anarrative proposal, but has become the past. This makes it impossible,

History as deconstruction 73

in e!ect, to distinguish between language-use and past reality. It is atthis point that empiricism notches up another success.

What is undeniable is that it is historians who construct narrativesthrough which historical knowledge is acquired and disseminated. Howis it possible for us to distinguish between the narrative proposals ofdi!erent historians, between those likely to be right and those wrong?How can we tell good history from bad history? This is not too hard todo for reconstructionists. They judge the degree to which the narrativelacks structure, unity and/or coherence in its congruence or corres-pondence with its contextualised sources. The most convincing his-torians are those who write narratives possessing this in full measure.Unity and cohesion are found in the intelligible and reasonable rela-tionship established between individual statements and the sources, buteven more importantly the narrative as a whole possesses an informingstructure of argument – the article, essay or book is not wa#y or ram-bling. In ‘good history’ the informing narrative argument will contain aclear and up-front statement as to how the past actually was – thecoherence of form coming from the overarching social theory deployed,or the fact that they have got the story/theory straight according to theevidence.

What is ‘good’ or ‘bad’ history for the deconstructionist? Hopefullythe narrative in a deconstructionist essay will be coherent and sensible,but it will not be epistemologically self-assured. This lack of certaintyarises because of the doubts harboured about correspondences. Howcan we readily di!erentiate truth-e!ect plausibility from fact? How maywe disentangle social theory arguments from low-level descriptions ofevents? How can we unpick ideologically inspired gaps and silences orunravel the collapsed signifier–referent? For every history that aims toget at the past as it actually happened, there is always another version,which, like the first, is by definition another fiction. As to what consti-tutes good history, then, it is that which is self-reflexive enough toacknowledge its limits, especially aware that the writing of history is farmore precarious and speculative than empiricists usually admit. Decon-structionist history openly accepts a dissenting role for the historianas someone who must challenge the established notions of authoritywithin contemporary society by refusing to ‘tidy up’ the past by ascrib-ing origins and causes with the claim to evidentially certified truth.What does this mean in more practical terms, and what are its implica-tions for history as narrative?

We have now arrived at two conclusions about history: first, all com-posed, written narratives are supported by a philosophy or ideology,often buried so deeply that no amount of conscious historical aware-

74 History as deconstruction

ness can eliminate it; and second, because it relates stories about realpast events in the evidence, deconstructionist history is not a fictionalnarrative. But, as a form of representation, all historical narrative pro-posals are shaped by the conventions of rhetoric and language-use –emplotment, argument and other culturally provided constraints, bothmaterial and ideological. This relationship between narrative form andhistorical content is explored by Hayden White in his study of historicalinterpretation, which in turn owes much to the investigation of lan-guage and representation undertaken by Roland Barthes, Paul Ricoeurand Michel Foucault.46 For the anti-narrativist White, the essence ofhistory is that it is a literary enterprise, and we ‘know the past’ throughthe narrative design we impose on it, which, as Ankersmit agrees,‘acquires a substantiality of its own’.47 Both White and Ankersmitrequest that before historians can embrace the true character of histori-cal explanation through a figurative narrative, we must resist the temp-tation of keeping up our pretence to objectivity, and turn instead to aricher understanding which is to be gained through an appreciation ofhistory as literature. In his 1973 text Metahistory, White argued that allhistory writing is basically a linguistic and poetic act. Facts are notdiscovered, they are actually sources interpreted according as much toliterary as any other criteria. Consequently, if we approach history asliterature we may even write better history, as we deploy an additionalrange of critical apparatuses to the established rules of contextualisedevidence. By recognising its literary form we are not constrained topresent it as mainstream history would have it done.

While the new empiricists continue to argue about how to comprom-ise ‘what happened’ with history as a representation of something thatdoes not (no longer) exists, there are ‘experimental historians’ who arewilling to explore a di!erent kind of history – ‘unconventional his-tory’.48 Since the advent of the ‘aesthetic turn’ in the mid to late 1990s,such experimental historians have bracketed o! the classic empirical-analytical method of ‘telling it as it really was’, in favour of a di!erentapproach to history. The whole point of non-experimental or ‘properhistory’ is not to introduce any epistemological doubts about the realityof the past as represented in the history text. But, as Brian Fay in hisintroduction to the themed issue on ‘Unconventional History’, in thefirst issue of the journal History and Theory in 2002, said that a goodreason for examining unconventional history was to become more clearas to what conventional history looked like. Alternatively, doingunconventional history may open avenues to new ways whereby we canunderstand the past as the process of making history. To do that, ofcourse, requires a brave step beyond the conventional.

History as deconstruction 75

This means rethinking the nature and forms of the representation ofhistory. Indeed, the British-American edited Rethinking History: TheJournal of Theory and Practice began publication in the mid-1990s toconfront the epistemological approach to history and conventional his-tory practices. It remains unique in this respect. The journal, and thenotion in particular of experimental history, has been met with a var-iety of responses, each di!ering according to the epistemological pref-erences of and emotional e!ect upon individual historians. The notionthat if we cannot recover the story of the past, then we should be able to‘experiment’ with di!erent stories, upsets all reconstructionists, it dis-mays the vast majority of constructionists, and is generally welcomedby deconstructionists. The reaction to experimental history seems to bein inverse proportion to the veneration in which the individual historianholds empiricism. And, maybe, how much they value their place in thehierarchy of the academy, especially in Britain. The notion of experi-mental history faces o! the misinterpretation of post-structuralist/postmodernist history that nothing of any value can be written aboutthe past.49 Indeed, the logic is quite the reverse. Perhaps the most valu-able things we can write about the past are those that emerge fromrecognition of its representational frailties and epistemologicaluncertainties.50

If we accept that the distinguishing feature of our postmodern condi-tion is epistemological scepticism, it is truly old news to say that historyis ‘positioned’ and that it cannot reflect the past ‘as it actually was’.Knowing ‘what happened’ does automatically, via language or anyother tertium quid (third mechanism), connect ‘the past’ with ‘history’.For history, the matter of the truth of the meaning of the past comesdown to how we represent the content of the past – our evidence – andhow the form of that representation constrains what we feel justified inbelieving what that evidence means. This situation makes the idea oftruth a great deal more complicated. So, the idea that the past is stillreally there in the sources and through them is only reinterpreted as newsources ‘are discovered’ is not merely implausible but such a beliefcloses o! our thinking about our ontological acts of thinking/rethinking and writing/rewriting.

This can be seen vividly in the new forms of history that are availabletoday. Experiments with history tend to concentrate upon the forms ofexpression in which they are cast. These normally di!er substantiallyfrom the normal modes of expression that are professionally agreedand sanctioned. These forms are driven by a realist epistemology inwhich ‘the past’ signifies itself as ‘history’. The meaning of this signi-fier–signified relationship is the substance we give to it as a history

76 History as deconstruction

lecture, or a history book or article and, because it is a professional riteof passage, the history PhD. Hence, these approved forms reflect theepistemological assumption that the historian gets the facts straight atone level and thereby discovers the history at another. As a result, non-sanctioned forms of expression that allow for empirical and analyticaldeviationism are seen as degraded and non-professional. Exampleswould include film, hyper-text, television history, the graphic novel andthe comic strip. Forms that appear to be outlandish modes of expres-sion, such as dance and historical re-enactment, are epistemologicallyinadequate and beyond the realm of history. It is suggested such formscannot be ‘proper’ history because they are performative and notempirically and analytically acceptable. We are now confronted with awholly new issue. This is how the historian-as-author connects the con-tent of what happened in the past with the form we give to it as history.This is the central issue in historical work today. So, experimentalhistory is a history that is, as Hayden White has long argued, self-conscious about its own narrative logic. It not only acknowledges itselfas a site under construction, but examines its construction as the essen-tial feature of the meaning it has.

Because written history is a literary artifact, White claims that his-torians share the same formal narrative structures used by writers of real-ist story literature based on the main categories of figurative language –the tropes – what White calls tropic prefiguration. White uses some-thing like a base-superstructure metaphor himself to explain how thisworks. Historians construct narratives (stories) to produce explanationsemploying three superstructural strategies of explanation, viz., explan-ation by emplotment, explanation by formal argument and explanationby ideological implication. These strategies of explanation are the sur-face features of the narrative, with White suggesting a deep or infra-structure of consciousness (operating at the level of the tropes) thatultimately determines how historians elect to explain the facts exploredin their narratives. Extending the base-superstructure metaphor, Whiteargues that language is not to be located in the economic base ofsociety, nor the social superstructure, but is prior to both.

Next, White carries forward the analysis from the level of rhetoric tothat of the historical by borrowing Michel Foucault’s concept of theepisteme – a way of describing how a culture in each age acquires anduses its knowledge as embedded in figurative language. White suggeststhat it is possible for historians to interpret the culture of any historicalperiod with reference to its ascendant tropic prefiguration.51 Whiteproposes that as the tropes organise the deep structures of humanthought in de Saussure’s sense of constituting meaning through binary

History as deconstruction 77

opposition – the idea of otherness, or di!erence in any historical period– tropes lie at the core of every society’s and every historian’s historicalimagination.52 White has explored the literary theory of tropes as a wayof distinguishing the dominant modes of the historical imagination innineteenth-century Europe, and by extension to the cultural level hismodel allows the identification of the deep and surface structures of thehistorical imagination.

I will explore the importance of this view of history in more detail inChapters 7 and 8, but for the moment it is important to note that thekey to this narrative model of cultural change is White’s conjecture thatideology and the exercise of power are ultimately settled by the cardinaltext, yet operate in the real world of social relationships.53 In movingfrom the rhetorical level to that of the material context, White isdescribing the writing of history as an intertextual and material act,with history as a conforming or dissenting voice. This he attempts todemonstrate in his analysis of E.P. Thompson’s Making of the EnglishWorking Class, claiming that like all history it is a necessarily fabricatedwork because of its inevitable dependence on the tropic model of his-torical explanation. Thompson is in the business of metaphorically‘making’ the English working class for overtly ideological reasons.According to White, ‘The pattern which Thompson discerned in thehistory of English working class consciousness was perhaps as muchimposed upon his data as it was found in them.’ But White goes on tomake an even more telling point: ‘the issue here surely is not whethersome pattern was imposed, but the tact exhibited [by Thompson] in thechoice of the pattern used to give order to the process being repre-sented’. As White says, the ‘planned or intuitive’ tropological patternThompson selected for the English working class is the movement froma ‘naïve (metaphorical) to a self-critical (ironic) comprehension ofitself.’54 What is significant for historians in White’s analysis of history ishis questioning of the relationship between the trope and social andcultural practice. In his work Mythologies, Roland Barthes also inter-prets language as being assembled by one social group to be consumedby another as ideology.55 With others like the anthropologist Cli!ordGeertz and cultural critic Michel Foucault, White has constantlyreviewed the representational and ideological status of the tropes(metaphor) in forming the social institutions of power andconsciousness.56

White is fully aware of another central problem raised by his rhet-orical approach to the study of history, and that is the fear of extremeinterpretative relativism. This can threaten a ‘free play’ of interpretativefantasy that may take us further from, rather than closer to, the origin

78 History as deconstruction

and subject of the evidence. White accepts that we have here a divisionbetween the historian who wants to ‘reconstruct’ or ‘explain’ the pastand one who wishes to interpret it or use it as ‘the occasion for his ownspeculations on the present and future.’57 In following Foucault’s logicon the text–context relationship, White does draw a line at the argumentof Jacques Derrida that there is only figuration and hence no meaningin and through language.58 The deconstructive historian need not betrapped in a forlorn snarl of rhetorical relativism. White believes, alongwith Foucault, that we can actually know many things about the realworld despite the limitations of language. But withal there remains hiswarning about the power of language:

The use of a technical language or a specific method of analysis,such as, let us say, econometrics or psychoanalysis, does not free thehistorian from the linguistic determinism to which the conventionalnarrative historian remains enslaved. On the contrary, commitmentto a specific methodology . . . will close o! as many perspectives onany given historical field as it opens up.59

The charge of rhetorical relativism, with its descent into moral declineand the sink of ideology, is countered by White’s claim that all lan-guages – whether the language of supposed objective history, or of thepoet – are equally relativistic, and equally limited by the languagechosen ‘in which to delimit what it is possible to say about the subjectunder study’.60 When the historian interprets the past he/she is notinventing it, or producing a fictionalised version that plays with the realevents and real lives of the past. The historian is rather imposing anarrative structure that has coherence and unity, endowing the past‘experience of time with meaning’.61 It is far from a descent into rhet-orical relativism (and the moral turpitude, as Saul Friedlander suggests,that would deny events like the Holocaust) to recognise that the past isintervened in when emplotted by historians, or, as Ricoeur puts it, ‘thenarrative art [that] characteristically links a story to a narrator.’62 WhatWhite is saying is that it is the function of the historian to explore theemplotments that may already exist in the past:

The meaning of real human lives . . . is the meaning of the plots . . .by which the events that those lives comprise are endowed with theaspect of stories having a discernible beginning, middle, and end. Ameaningful life is one that aspires to the coherency of a story with aplot. Historical agents prospectively prefigure their lives as storieswith plots.63

This daring vision of the historical enterprise necessitates rather than

History as deconstruction 79

denies the kind of attention to the evidence that all empiricists andcontextualists would applaud. The logic of this argument is that wehistorians, while we tell stories, have little of the imaginative freedomexercised by writers of fiction because we are in the business of theretrospective emplotment of historical events and narratives. While thehistorical account is a figurative exercise in the sense of being a productof the literary imagination, its relativism remains limited by the natureof the evidence.

CONCLUSION

The deconstructive consciousness raises several fundamental questionsabout the character of history defined as the reconstruction of the pastaccording to the available sources, and the construction of the past bythe imposition of explanatory frameworks. The empiricist argumentthat our knowledge of the past is derived through the painstakingstudy and interpretation of fragmentary and partial evidence, and thatthe sheer professionalism of the working historian will overcome theproblems of bias, ideology and the many other obstacles to historicalunderstanding, is countered by the proposal that history is instead arecognition of the intimacy existing between content and form. In otherwords, we remind ourselves that history is not only about the sifting ofevidence and constitution of facts, and that interpretation itself is anact of linguistic and literary creation.

This approach to historical analysis suggests that that which we call‘the historical’ cannot be understood in all its fullness by a priori logic,positivism, or by the painstaking reconstructionist analysis and consti-tution of facts alone. Instead we may grasp more of the richness ofhistorical analysis by incorporating into the study of the past the inter-textual nature of history as a discourse. The truth found in histories,White suggests, ‘resides not only in their fidelity to the facts of givenindividual or collective lives’ but ‘most importantly in their faithfulnessto that vision of human life informing the poetic’.64 It is by recognisingthe expressive and figurative content of historical narrative, ‘the contentof its form’,65 that the historian contributes to our understanding of thepast. This does not mean that we historians only examine the purelyfigurative or metaphorical level of the discourse of history, but weintervene in the past by actively extrapolating from the literal to thesymbolic level of understanding, from the present to the past.

Perhaps the central point about the deconstructive turn is the recog-nition that narrative upsets the assumed balance between language andreality. Historical language (Ankersmit’s narrative proposal) becomes

80 History as deconstruction

the primary vehicle for understanding. We should abandon the trad-itional empiricist epistemology in favour of a radical new hermeneuticor interpretative approach to the generation of knowledge about thepast. I will elaborate on this significant suggestion later in my moredetailed study of Foucault and White. For now I will repeat that wemust examine the figurative use to which the historian puts the literalsense of meaning he/she has supposedly discovered in his/her research.This applies not only to the interpretations of historians but also to oursources. Consequently, every history is always something more than theevents described. The historian represents the past rather than reclaimsit as it really was. It is the deep suspicion, generated by this emphasisupon narrativisation and presentism, that motivates the empiricistcritique of the deconstructive consciousness. Deconstructionists, it isclaimed, forget the sources, the problems of research, and assume thatideology must unavoidably colour our historical descriptions. It is tothis critique of deconstructionist history that I now turn.

History as deconstruction 81

5 What is wrong withdeconstructionist history?

INTRODUCTION

The idea of meaning being located in a narrative or representationmodel of historical explanation is, for conservative reconstructionists,as much a constructionist-type imposition as is explanation throughsocial theory. But it is not just hard-core reconstructionist historianswho reject so-called postmodern history: a broad group of pro-narrativist practical realists such as Frederick A. Olafson, James Klop-penberg, James Winn, James F. McMillan, Joyce Appleby, Lynn Huntand Margaret Jacob also seriously doubt the kind of history promotedby the deconstructive consciousness. In a summary of their position,Olafson insists that ‘It is not . . . possible to give up all truth claims . . .for historical interpretations.’1 As I have tried to show, deconstruction-ist history confronts each of the six principles of traditional empiricisthermeneutics under each of the four headings of epistemology, evi-dence, social theory and narrative form. The message of the decon-structionist consciousness – that the mainstream still pursue past realitythrough their assumption of an objective study of the sources – isrejected by those who argue that this image of what historians do todaymisconceives and grossly oversimplifies the nature of traditionalhistory. So, from the established perspective, what is wrong withdeconstructive history?

EPISTEMOLOGY

In the view of the British historian John Tosh, there is at one extreme ofthe profession those like G.R. Elton ‘who maintain that humility in theface of the evidence and training in the technicalities of research havesteadily enlarged the stock of certain historical knowledge’,2 but thatmethodologically to be a historian means that most ultimately do

incline towards Elton’s position. Those that do not, like TheodoreZeldin, who insists that no historian can o!er more than a personalperspective on the past, are certainly not in the two main tendenciesof accepted practice.

As moderates, Appleby, Hunt and Jacob have it that the postmoderndethroning of independent reality perceived as objective historical truthhas been a di"cult process of discovering ‘the clay feet of science’.3 Asreasonable historians of goodwill describing themselves as practicalrealists, they suggest that the Second World War and subsequent ColdWar together produced a substantial scepticism about the verities ofscience and what constitutes truth. Our present age of uncertainty wasalso consequent on the mid-twentieth-century Kuhnian–Popperiandebate over how science gets at truth. Thomas Kuhn, arguing in favourof so-called paradigm shifts whereby science suddenly challenged andtransformed its dominant theoretical constructs, seemed to be openingup scientific proof to the influence of social forces.4 This apparent end-ing of objective science was denied by Karl Popper, who argued that hewas no positivist because truth in science could be grasped only throughnon-falsifiable logical processes which might yield the covering laws ofhistorical constructionism rather than the messiness of empiricism. Notfor Popper the crudities of correspondence theory. The growing accept-ance by Appleby, Hunt and Jacob that knowledge may be socially con-structed does not mean of course that they accept that all truth isrelative or value-laden, nor less that writing about the past in the pres-ent (historicity) makes relativism inevitable. In fact, as Appleby, Huntand Jacob have it, the social reading of knowledge is actually epistemo-logically advantageous in the search for historical truth. Cautiouslythey declare that historical objectivity can actually emerge from, as theysay, the ‘clash of social interests, ideologies, and social conventionswithin the framework of object-oriented and disciplined knowledge-seeking’. For them, hard-won truth ‘however mired in time and lan-guage’ is still truth in a democratic society.5 The issue of truth is usuallyclaimed to lie at the heart of the nature of history. Without the notionof history as an empirically cognitive instrument, then it might as wellbe fiction. Empiricism, after all, is the only thing that is supposed todistinguish history from fiction. Historians have to rely on knowingas precisely as possible what happened. This is their epistemologicalanchor. The issue is not, unfortunately, quite so straightforward. To putit at its plainest, the deconstructionist argument is that knowing whathappened does not tell you what it means. The provision of meaning forthe past requires a far more complex mechanism than simply ‘gettingthe story straight’. For a start, is there ‘the story’ to get straight? The

What is wrong with deconstructionist history? 83

notion of ‘the story back there waiting to be discovered’ requires a hugenumber of assumptions, many of which are really very doubtful. Ofcourse, it should be mentioned that no deconstructionist argument canclaim (like any other) to be authoritative. In other words, the analysis oftruth in history from a deconstructionist perspective has no more claimto be the truth about truth than any other.

That necessary reminder having been issued, the first assumptionmade by empiricists is that the evidence is by its nature capable of beingknown for ‘what it really means’. A central plank in this belief (that it isknowable for what it means) depends on assuming that the historiancan know the intentions of people who acted in the past. There hasbeen a substantial debate on this over the years. Most recently, thedispute – in e!ect between the extremes of objectivism and relativism –received an airing in an exchange between Mark Bevir and Frank R.Ankersmit.6 It seems common sense to work from the principle that wecan know what a text means if we reasonably assume the text meanswhat the author intended it to mean. Unfortunately, as Ankersmit andothers have pointed out, the context and the language conventionsdeployed in the text, rather than simple authorial intention, usuallygovern the meaning of a text. This suggests that meaning does notalways coincide with intention. Authors can say one thing and meananother. Do intentions cause meanings? Indeed, where do we locate thewriter’s intentions? We can’t. The reason is that few, if any, writers inthe past (or present?) o!er commentaries explaining what they meantto say. But even if they did, the basic problem, clearly, is not resolved.Explanations about statements that the author has made leave us nobetter o!. The problem of knowable intentionality will never evaporate.

The second assumption about truth is that truth in history is a matterof the comparison between ‘past reality’ and ‘history’. This is, as I havesuggested, a very doubtful belief. Instead what we have to do, asAnkersmit and Bevir suggest, is compare historical representations witheach other. Since we cannot ‘go back’ to the past to judge it against ourhistory, we can only measure history against history. That we have ref-erence in the history is entirely irrelevant. Huge wedges of footnotes(though always desirable for reasons other than guaranteeing meaning)are quite immaterial beside the issue of not being able to comparehistory with the past. Like lawyers in this respect, we can only compareaccounts of what happened for a basic agreement that someone isn’tinventing what went on. But once this important though routinerequirement is satisfied, we are still no better o! in terms of what it allmeans. It might be di!erent, perhaps, if we could actually relive orre-experience first-hand the past. But even then, we would face other

84 What is wrong with deconstructionist history?

problems, such as failing or uncertain memory. And, of course, wewould still have to cast our relived experience as a narrative and thatbrings us right back to the start of the problem again. It might well turnout to be unfortunate if the empiricist desires high degrees of certaintyin what they think about the past, but that is not the happy situation inwhich historians live.

The postmodern age of which deconstructionism is an attribute isthus keynoted by an enhanced and unwarranted sense of irony at thepresumed disappearance of the certainties of objective knowledge, inwhich disciplines are viewed as historical cultural practices (meta-narratives) or canons intended not to generate truth and unbiasedknowledge, but rather to sustain present or prospective dispensationsof dominance and subordinance. From the postmodern or decon-structionist perspective objective science is not objective, non-sectarian,universal and transcendent, but it legitimates the present dominantforms of Western civilisation. Lyotard, Foucault, Barthes and Derridahave argued that we are unable to represent reality accurately in lan-guage, and that we cannot, therefore, assume that objectivity is feasible,nor should we accept either the correspondence theory of knowledge,or Popper’s logical positivism. It follows that the notion of theindividual as a potentially autonomous, non-ideological animal isalso flawed, and knowledge produced by such a creature must be afabrication, an assembled invention that disguises the will to power.Hence, the wedge of relativism and scepticism reaches its apogee inpostmodernism.

Such a Nietzschean-inspired universe which has no truth, whichaccepts the failure of representation, and which consequently acceptsrelativism in moral standards, is rejected by traditional history.Foucault’s position is likewise rejected where it means accepting ananti-humanism that itself rejects human agency.7 By the same token, thetraditional paradigm has it that history is quite capable of recognising‘the other’, the marginal, the oppressed and the hegemony of bourgeoisideology that sanctions colonialism and Third World exploitation. So,how is it possible to base historical method on the belief that there is nodirect access to independent knowledge because there is no clear separ-ation between objectivity and subjectivity, fact and value, and historyand fiction and truth can never be more than perspective? Appleby,Hunt and Jacob summarise the deconstructive perspective asassuming that

Human beings do not achieve a separation from the objects theystudy; they simply invest them with their own values. Thus along with

What is wrong with deconstructionist history? 85

modern history, the idea of the human being as an autonomous,subjectively willing, rational agent [is] brought into question.8

What is being questioned by Appleby, Hunt and Jacob’s traditionalhermeneutics is the inevitable result of the Saussurean position that iflanguage is built on the arbitrary relationship of signifier and signified,we must doubt the correspondence theory of reality and deny thecommonsense notion that the truth is ‘out there’. Although the prac-tical realist mainstream can accept, along with deconstructionist his-torians, that the reality of the past is always mediated through culturallyinfluenced narrative structures, that language is not a transcendentmeasure of truth and, as another American historian, Linda Gordon,has concluded, that objectivity is ‘certainly an issue’, deconstructionistrelativism is not the only response.9

The most rugged rebuttal of the deconstructive turn has undoubtedlycome from the British Tudor historian Geo!rey Elton. Elton has strik-ingly combined his long-standing opposition to constructionism andwhat he views as E.H. Carr’s scepticism, with an equally energeticrebu! to deconstructionist historical analysis. Quoting correspondencetheory, Elton insists that historical truth is the product of the indepen-dent relationship between knower and known, and historians aspire toreach it, but any failure to achieve it on their part ‘does not abolish thetruth of the past event’.10 With regard to the historian seeking truth, thephilosopher of history Michael Stanford echoes Elton in claimingthat the historian ‘is permitted only one attitude – that of impartialobserver, unmoved equally by admiration or repugnance. Nor does hepresume to dictate the reader’s response; he simply relates the facts’.11

Arthur Marwick supports Elton’s position with his own attack on‘postmodern relativist’ Hayden White. Marwick sustains the notion ofhistory as a distinct epistemology by denying White’s ‘imagined dis-junction between what the historian discovers . . . and the writing up ofthose discoveries’. According to Marwick, White produces a ‘rabbit-out-of-the-hat’ with his claim that all written history must ‘obey thecodes of narratives and discourse’ and consequently cannot ever belogically demonstrated. Marwick is firm in the argument that historyexists independently of the historian and is certainly not written accord-ing to White’s formal grid or structure of tropes and emplotments.Marwick remains convinced that the historian is not ruled by structuresof language to the extent that historical truth can never be known,concluding that ‘ “deconstruction” and “discourse analysis” . . . are nouse to historians looking for precise, and in some sense unique, answersto specific questions’.12

86 What is wrong with deconstructionist history?

This staunch Elton–Stanford–Marwick defence of fundamentalistreconstructionist epistemology also finds expression in the mainstreamvoices of Appleby, Hunt and Jacob in their more nuanced but stillultimate denial of the cognitive value of the written form of history.First they pose the fundamental deconstructionist question of ‘howdoes the historian as author construct his or her text, how is the illusionof authenticity produced, what creates a sense of truthfulness to thefacts and a warranty of closeness to past reality (or the “truth-e!ect” asit is sometimes called)?’. Their answer is that history is unmistakably anindependent discipline and not merely a mimetic or ersatz literary genredependent for its power to explain on the trompe l’œil principle.Appleby, Hunt and Jacob regard the deconstructive argument as beingproductive of not only relativism, but of a relativism ‘possibly tingedwith cynicism or arrogance’ that insults the e!orts of people in the pastwho themselves believed that they were seekers after truth. Finally,without the ability to ‘represent reality in any objectively true fashion’,they say that we can never expect to explain anything at all.13

A further defence of empiricism as the foundation of traditionalhermeneutics is o!ered by American historians like James Winn, LindaGordon and especially James T. Kloppenberg.14 Relativism, or ‘intel-lectual and moral chaos’ as Peter Novick described it, is not the onlyalternative to objectivity. Like Appleby, Hunt and Jacob, James Klop-penberg fortifies reconstructionism by invoking certain arguments ofcontemporary pragmatic philosophers Richard Rorty and Richard J.Bernstein.15 Instead of accepting the six key principles of empiricism asinevitable dualisms, Kloppenberg attempts to moderate the absolutismof empiricist objectivism. He does this by suggesting that history can beregarded as epistemologically viable through a ‘pragmatic theory oftruth [that substitutes] continuing social experimentation for certainty[accompanied by] a historical sensibility’ that conceives of all knowl-edge ‘as intrinsically meaningful and rooted in cultural processes thatcan be known only through interpretation’. This defence of empiricismdoes away with Derrida’s extreme insistence on relativism as the inevit-able result of the collapse of objectivity. Consequently, history remainsa legitimate epistemology because of its fundamental empiricist deduc-tive – inductive methodology. As Kloppenberg says, articulating themainstream methodological position,

Hypotheses – such as historical interpretation – can be checkedagainst all the available evidence and subjected to the most rigoroustests the community of historians can devise. If they are verifiedprovisionally, they stand. If they are disproved, new interpretations

What is wrong with deconstructionist history? 87

must be advanced and subjected to similar testing. The process isimperfect but not random; the results are always tentative but notworthless.16

From this perspective (derived from the first two of the six key prin-ciples), the duality proposed by deconstructionists of historians aseither objectivists (naïve realists) or relativists (sophisticated ironists)loses its validity. Kloppenberg’s is not a new response. The Americanhistorian Charles Beard (with Carl Becker) argued in the 1930s thathistory was then a complex process of ‘hermeneutics and pragmatictruth-testing, in which knowledge derives from weaving together factand interpretation to create stories [myths, Becker called them] whoseaccuracy must therefore always be considered provisional’.17 Feardaccepted as incontrovertible the principle that facts could be acquiredobjectively, but recognised that absolute truth as a historical general-isation claiming to explain those facts was an unobtainable goal. As heoften said, ‘We see what is behind our eyes.’18

Reconstructionists, ranging from Elton, Stanford, Marwick andHimmelfarb to mainstreamers like Kloppenberg, Appleby, Hunt andJacob, are thus agreed on defending history as a distinct epistemology.Where they divide is on the historian’s use of models, the deductivemethod and use of narrative. Hard-liners defend the discipline’s induc-tive methods on the grounds o!ered by Leon Goldstein that deductionhas ‘no role at all in the . . . road to historical truth’,19 while practicalrealists would join the conservatives in their opposition to deconstruc-tion because of their shared belief in the ultimate existence of a know-able reality ‘out there’ located in the evidence, and which those soinclined can interrogate through their social theorising. It follows that itis how we treat the evidence that in large part determines the responseto the deconstructionive consciousness.

EVIDENCE

In the previous chapter I suggested that deconstructionists question theauthority of the source in several ways: by maintaining that the inten-tion of the author of the evidence must always remain unknown (thedeath of the author), by considering the understanding of evidencethrough its contextualisation to be a doubtful procedure, and bycasting doubt on the explanatory power of the practice ofQuellenkritik. The majority of historians today can be described aspractical realists who acknowledge that the actuality of the past isalways imperfectly encountered, whether through social theory or

88 What is wrong with deconstructionist history?

narrative impositionalism. Those practical realists in the mainstreamtendencies like to argue that today only deconstructionist historians gethot under the collar about endless chains of signification and a realitythat is always in an altered state, given its narrated form. They arguethat deconstructionist colleagues have erected a straw man in their over-inflation of the issue of crude empiricism – as Marwick has suggested,‘postmodern critics . . . totally misconceive the way in which historiansgo about their business’.20 The problem of indeterminate meaning isnow openly acknowledged. The only di"culty lies with the decon-structionist refusal to understand that most mainstream practicalrealist historians today accept their role as being essential to theinterpretation of written evidence, rather than demonstrating thefailings of empiricism. They point to E.H. Carr, who said almost fortyyears ago that facts emerge as historical facts only when processedthrough the mind of the historian. Historians are by definitioninterpreters and not just facilitators of meaning.

Allowing for the deconstructionist arguments that language cloudsrather than clarifies meaning, that there are a multiplicity of meanings inour source texts and that the author of the source is (like the historianalso) the creature of multiple cultural discourses and significationsdoes not mean that culture and ideology write history. Even grantedthese constraints, reconstructionist historians would argue that neitherhistorians nor sources are adrift on a sea of significations nor necessar-ily subject to the tides of cultural relativism and ideology. No reason-able historian today, or quite possibly ever, has claimed either thatempiricism is a system that guarantees the objective discovery of truth,or that there can ever be a hermetic seal between knower and known.As John Tosh has said, the process of creating historical knowledgestarts with the questions the historian ‘has in mind at the outset ofresearch’.21 This is natural, normal and nothing to worry about. Mar-wick adds, ‘the technical skills of the historian’ lie in sorting out theproblems with sources as well as deciphering the ‘codes of language’they employ.22

The nature of the access to an independent and real past as repre-sented in the evidence has been extensively addressed by C. BehanMcCullagh and Frederick A. Olafson. McCullagh holds that the basicyardstick for historians is the commonsense and commonplace assump-tion that their perceptions ‘are caused by roughly similar states of theworld . . . an assumption which we all make most of the time, withcomplete equanimity and success’.23 Founded on the assumption thatthe world exists separate to our knowledge of it, historians ‘do notconstruct past reality by attempting to describe it. All that they

What is wrong with deconstructionist history? 89

construct is our knowledge of it, our beliefs about it’.24 As a result,historians constitute the truth of historical descriptions not by a directcomparison with the reality of the past, because this, it is agreed, isinaccessible, ‘but by inferring them from the present evidence’.25 Decon-structionists are thus presumed consistently to underestimate thesophistication of the inferential method. They miss the genuine sophis-tication and complexity of historical practice today. Olafson declaresthat the evidence of the past, as found in the statements of the author(s)of the evidence, is directly referential in so far as the evidence refers tothe past. Evidence thus construed does actually signify a referentiallinguistic event. Inductivism – inference from the sources, or Quellenkri-tik – thus remains the primary defence against the deconstructionistperspective. Of course, inductive inference can be wrong if based onfalse or non-verifiable evidence. Consequently the reasonable historianalways distinguishes what he/she believes to be true about the past, andwhat may actually be true. Such a belief is inevitably founded uponsome idea about what the world of the past was like and how itwas ordered. This initial thought is the start of the process for theconstruction of historical facts.

This, the most popular vision of the use of evidence in history, isderived from E.H. Carr in his highly influential 1961 book What isHistory?. By following the logic of R.G. Collingwood (a historianscorned for his relativism by Marwick and Elton), Carr set aboutanswering the question ‘What is a historical fact?’. He argued that his-torical facts are derived through ‘an a priori decision of the historian’.It is the manner in which the historian arranges the facts derived fromthe evidence, as determined by his/her prior knowledge of the context,that creates historical meaning. Using Carr’s analogy, a fact is like asack: it will not stand up until you put something in it.26 The somethingis a question addressed to the evidence. However we describe it, his-torians quite legitimately impose on the past through their knowledge-based interrogation of it. As Carr insists, and mainstream historianswould agree, ‘The facts speak only when the historian calls on them: it ishe who decides to which facts to give the floor, and in what order orcontext.’27 In The Idea of History, his earlier study of the creation ofhistorical meaning, Collingwood sugggested that the historian arrangesthe information available about the past in the light of the context,which he described as a ‘web of imaginative construction’.28 Factsare constituted when they are verified by comparison and placed in ameaningful relation to each other in the overall historical context.

As Carr describes the derivation of the historical fact, ‘Its status . . .will turn on a question of interpretation. This element of interpretation

90 What is wrong with deconstructionist history?

enters into every fact of history.’ He concludes, much to the vexation ofElton, that the ‘historian is necessarily selective. The belief in a hardcore of historical facts existing objectively and independently of theinterpretation of the historian is a preposterous fallacy, but one that isvery hard to eradicate’.29 Since the 1960s Carr’s arguments have consti-tuted the dominant paradigm for moderate reconstructionist historiansbecause he pulls back from the abyss of relativism to which his ownlogic, and that of Collingwood’s, directs him. In the end Carr rebu!s Col-lingwood’s excessive insistence on the formative role of the historian,and replaces it with an image of the historian who, in acknowledg-ing the dialogue between past events and future trends, believes that asort of objectivity can be achieved. This is not then the absolutism ofthe crude Eltonian reconstructionist, but a workable or pragmaticobjectivity based upon a degree of self-reflexivity – a position fullyendorsed by Appleby, Hunt and Jacob.30

Facts remain, then, a contested terrain, and it is not something newthat deconstructionists have discovered. There are, of course, somehard-core empiricists like Peter Gay, who would deny Carr’s relativistconclusion while accepting that his thinking, ‘the historian’s mental setor secret emotions’, rather than producing distorting interpretations,may actually provide a clear view of the past.31 As Gay says, ‘To equatemotive with distortion . . . is demonstrably illegitimate’, especially if themotive drives ‘the inquirer toward the e"cient comprehension of theoutside world. The need that generates inquiry may be sublimated intodisinterestedness. Even empathy, the very emotion that the modern his-torian is ceaselessly enjoined to cultivate, has its objective component.’Gay concludes, in support of what Novick calls the ‘hyperobjectivistposition’, that although the rhetoric of history is di!erent to that ofscience, ‘this does not entail the expulsion of history from the family ofthe sciences. It simply makes the historian’s science special, with its ownway of telling the truth’.32 In this fashion Gay provides perhaps theultimate denial of the deconstructive turn – interventionist historianscan write objective history.

The motivation behind the work of the historian is found in thequestions they ask of the evidence, and it is not, as deconstructionistswould have it, automatically to be associated with ideological self-indulgence. Deconstructionist worries about motivation and inferencefrom the sources have little strength as long as historians do not pre-conceive patterns of interpretation and order facts to fit those pre-conceptions. As McCullagh remarks, the fact that historians can locatemore than one pattern in the same evidence does not mean that thosepatterns cannot represent reality. Indeed, taking up Hayden White’s

What is wrong with deconstructionist history? 91

specific argument that there are many correct views of any object understudy, McCullagh regards this as old news and quite unproblematic,arguing that in virtually every case past events are ‘capable of severaldi!erent true descriptions’.33 Like most other practical realists, McCul-lagh is eager to challenge the deconstructive belief that historians areincapable of writing down even a reasonably truthful narrative repre-sentation of the past. He thinks that they can because their narrativesare based on a close scrutiny of the evidence.

Hence historians do not deny that the process of translating the evi-dence of the past into the facts of history involves making initial apriori interpretative decisions – that is part of the winnowing process asCarr calls it.34 This is right and as it should be. Although decon-structionists like Dominick LaCapra can bemoan fact fetishism, thisagain is old news. Almost half a century ago, Carr described the ‘fetish-ism of facts’ as a nineteenth-century fashion, not one that modernhistorians need concern themselves about. Deconstructionism, alongwith Elton’s conservative reconstructionism, underestimate the sophis-ticated nature of mainstream historians today in their dealings with theevidence in context. As John Tosh explains, they ‘seek to reconstruct orre-create it – to show how life was experienced as well as how it may beunderstood – and this requires an imaginative engagement with thementality and atmosphere of the past’. Moreover, ‘the evaluation ofdocumentary sources depends on a reconstruction of the thoughtbehind them’, and before anything else can be achieved ‘the historianmust first try to enter the mental world of those who created thesources’.35 It is up to the historian to turn sources into history. Sourcesare useful only when they are processed like raw materials into theevidence from which historical facts are created.

As Collingwood suggested, evidence does not constitute an o!-the-shelf or ready-made historical knowledge that can simply be swallowedand then disgorged by the historian. Source materials become useful ashistorical facts only when the historian has applied to them the rangeof contextualised knowledge he/she already possesses. It is not goodenough to rely upon the correspondence theory of historical proof, asdeconstructionists appear to believe historians do. The process of his-torical interpretation based on the evidence is far more sophisticatedthan the simple description of sources that deconstructionist historianslike to assume occurs. Neither do most historians accept a Popperianscientific model of explanation any more than they accept beingtrapped in ideology. The post-Kuhnian relativist and deconstructionistpreoccupation with the failings of empiricism and the illusory characterof objectivism are largely irrelevant because they address issues with

92 What is wrong with deconstructionist history?

which historians have long been familiar and with which inferentialhistorical methodology, if properly applied with an appropriate senseof practical realism, presents few genuine or insoluble problems.

From what I have said so far it should be clear how moderate orpractical realist reconstructionists work: through a complex deductive–inductive process that recognises history to be the product of a dialoguebetween historian and source. It is generally understood that thisrelationship involves hypothesis-framing, to the extent at least of estab-lishing preliminary interpretations or conceptualisations based onknowledge of the context as well as familiarity with the sources. Theseinitial thoughts are the essential first step when addressing a problemand/or new evidence. This is not a definition of constructionismbecause such thoughts have not reached the same level of activity asthat of social theorists, identified by Elton as ‘adherents of theory [who]do not allow facts to disturb them but instead try to deride the wholenotion that there are facts independent of the observer’.36 For main-streamers it is the bellicosity of such a conservative position that givesthe arguments of deconstruction more credence than they deserve andthat denies the importance of much highly valued constructionist his-tory. Most practical realists would argue that the volume and range ofhistory presently are testament to its vitality, and the ‘linguistic turn’and the influence of post-structuralist thought, which a voluble minor-ity of historians fear as a threat to the discipline, has instead given it anew lease of life as the ‘new cultural history’. It seems that history haslittle to fear from postmodernism, which is really only a diversion fromthe main agenda.

THEORIES OF HISTORY: CONSTRUCTING THE PAST

It seems clear that the line drawn between reconstructionist and con-structionist historians in their employment of a priori or deductivemethod is for practical realists a narrow one, but the debate overwhere that line is drawn has distinct implications for deconstructionisthistory. In an influential presidential speech before the AmericanHistorical Association in 1910, the social science-inspired Americanhistorian of the frontier, Frederick Jackson Turner, warned hisaudience that

The economic historian is in danger of making his analysis and hisstatement of a law on the basis of present conditions and then pass-ing to history for justificatory appendixes to his conclusions. . . . Thehistorian . . . may doubt . . . whether the past should serve merely as

What is wrong with deconstructionist history? 93

the ‘illustration’ by which to confirm the law deduced from thecommon experience by a priori reasoning tested by statistics.37

Heeding this sort of warning, Geo!rey Elton not only jettisons cover-ing laws but imagines something worse: an ardent pro-narrativist-inspired merger of speculative philosophy with a degraded empiricism.Mainstreamers generally consider that this is not the terrain on whichto enjoin battle with the deconstructionists.

As I have suggested, although Elton is dismissive of all theory, in hisconservative call to arms he reserves his most trenchant criticism fordeconstructionism, which he describes as ‘the conviction that since his-tory has to be written the only kind worth having operates within theframework of a general theory of language’, an idea that, along withvarious other kinds of constructionist impositionalism, undermines‘claims to rational, independent and impartial investigation’.38 What hecalls ‘the endeavours to use literary theory to destroy the reality of thepast’ can only do serious harm, like all other forms of constructionisttheory, to the historian’s first duty which is to reconstruct the past asobjectively and independently as possible.39 All that theory in historydoes, Elton maintains, is to turn the historian into its slave:

The theory directs the selection of evidence and infuses predestinedmeaning into it. All questions are so framed as to produce supportfor the theory, and all answers are predetermined by it. Historianscaptured by theory may tell you that they test their constructs byempirical research, but they do nothing of the sort; they use empir-ical research to prove the truth of the framework, never to disproveit. . . . Adherents of theory do not allow facts to disturb thembut instead try to deride the whole notion that there are factsindependent of the observer.40

Marxists, in a brief alliance with deconstructionist historians, reject thelogocentrism of the conservative bourgeois reconstructionist version ofa past ‘reality’. But Marxists eventually turn their guns on decon-structionism because it fails to acknowledge that texts, like beliefs andideas, are read and understood in the real world. For Marxists, thedeconstructive consciousness fails because it de-materialises reality.Marxists view deconstructionism as just another version of idealismthat de-couples human beings from their economic and social context.Texts have authors, even deconstructionist history has authors, and theintentionality of those authors can be seen in what they do, and read,and write in the material world of recoverable structure and pattern.

There are other fundamentalist reconstructionist apart from Elton

94 What is wrong with deconstructionist history?

who are concerned about both the constructionist and deconstruction-ist challenge to the Rankean paradigm. Their disquiet is well put by theAmerican social historian Gertrude Himmelfarb:

All historians, new and old . . . have something to worry about – notonly the fragmentation of history but the deconstruction of history –and not only on the part of avowed deconstructionists but on thepart of social historians who unwittingly contribute to the sameresult.41

For Himmelfarb, deconstructionism is merely a more perniciousversion of constructionism:

Although deconstruction, as a conscious, systematic philosophy, hasbeen most prominent among intellectual historians, the mode ofthought it represents, even its distinct vocabulary, is permeating allaspects of the new constructionist history. Historians now freely usesuch words as ‘invent,’ ‘imagine,’ ‘create’ (not ‘re-create’), and ‘con-struct’ (not ‘reconstruct’) to describe the process of historical inter-pretation, and then proceed to support some novel interpretation bya series of ‘possibles,’ ‘might have beens,’ and ‘could have beens.’42

For Himmelfarb, deconstructionism and constructionism are twosides of the same relativist coin. She reasons that the New History’s‘increased use of quantification, models and other social science tech-niques’ has produced not greater objectivity but ‘an increased sense ofrelativism and subjectivism’. Marxism is her regular target, but variousother kinds of social theorising are also threatening history:

It is not only political history that the new historian denies or be-littles. It is reason itself. . . . This rationality is now consciously deniedor unconsciously undermined by every manner of new history: . . .by anthropological history . . .; by psychoanalytic history . . .; byengagé (and enragé) history . . .; by the new history of every descrip-tion asking questions of the past that the past did not ask of herself,for which the evidence is sparse and unreliable and to which theanswers are necessarily speculative, subjective, and dubious.43

Such comments have been held to be ill-directed by Lawrence Stone,who suggests that Himmelfarb should not divide the mainstreamhistorical world at a time when the greater threat to rationality is notfrom constructionist New History but from ‘philosophy, linguistics,semiotics, and deconstructionism’.44

Stone had already declared his own anti-constructionist and decon-structionist position in the late 1970s when he claimed to have detected

What is wrong with deconstructionist history? 95

evidence ‘of an undercurrent which is sucking up many prominent“new historians” back again into some form of narrative’. Hecontinued:

In some countries and institutions it has been unhealthy that the‘new historians’ have had things so much their own way in the lastthirty years; and it will be equally unhealthy if the new trend, if trendit be, achieves similar domination here and there.45

While disclaiming any attempt to make value judgements on the newtrends, Stone explored the nature of ‘scientific history’ – translated asconstructionist Marxism, Annales, cliometrics and ‘other “scientific”explanations of historical change’ which ‘have risen to favour for awhile and then gone out of fashion’. Stone had both French structural-ism and Parsonian functionalism in mind. Instead of explaining thepast, Stone concluded that all these trends did was to usher in ‘his-torical revisionism with a vengeance’ as a result of their focusing on ‘thematerial conditions of the masses’ and relegating the major historicalmovements associated with the élite. As he said, ‘In this new model ofhistory such movements as the Renaissance, the Reformation, theEnlightenment and the rise of the modern state simply disappeared.’He finished with the thought that ‘This curious blindness was the resultof a firm belief that these matters were all parts of . . . a mere superficialsuperstructure.’ The revival of narrative was due to ‘widespreaddisillusionment with the economic determinist model of historicalexplanation’ and in particular with the Annaliste relegation of socialand intellectual developments. For Stone the way to reverse this processwas through a revival of narrative which would spin ‘a single web ofmeaning’.46

The description ‘revival of narrative’ describes Stone’s e!ort to movehistorical methodology away from what he saw as its constructionisteconomic monocausal determinism by groups of new historians nolonger constrained by a ‘specific methodology, structural, collective andstatistical’. By ‘narrative’, Stone was referring to a ‘cluster of changes inthe nature of historical discourse’47 which, in the 1970s, witnessed ‘aquite sudden growth of interest in feelings, emotions, behaviour pat-terns, values, and states of mind’. To this end he evidenced the work ofnarrative-inspired sociological theorists E.E. Evans-Pritchard, NorbertElias, Cli!ord Geertz, and political theorists J.G.A. Pocock and Quen-tin Skinner, whose ideas had been utilised by historians. For Stone, this‘movement to narrative by the “new historians” ’ spelled the end of theattempt ‘to produce a coherent scientific explanation of change in thepast’.48 It has to be pointed out, however, that for Stone ‘narrative’ as a

96 What is wrong with deconstructionist history?

term was a particularly bad shorthand to describe what was, in e!ect, abroad cultural reorientation among historians, especially because in the1980s the issue of narrative structure developed along more specificdeconstructionist lines.

In a 1991 issue of Past and Present, Stone criticised the latest post-modern trends in history which ‘brought seriously into question’ itssubject-matter, data and its mechanisms of explanation.49 Stone iso-lated three distinct threats from the constructionist and/or decon-structionist camps: ‘The first threat comes from linguistics, building upfrom Saussure to Derrida, and climaxing in deconstruction . . . Thesecond . . . cultural and symbolic anthropology. . . . The third . . . fromNew Historicism.’50 Stone maintained that together they challengedhistory’s basic empiricist principles. Stone elaborated on what he tookto be the essence of history today. Rather than the positivism of thenineteenth century which deconstructionists fondly believed was stillthe dominant practice, Stone established what for him were the keymainstream beliefs: that history should be written in ‘plain English,avoiding jargon and obfuscation’; that ‘historical truth is unattainable,and that any conclusions are provisional and hypothetical, always liableto be overturned by new data or better theories’; that we should acceptthat historians have bias and would do well, like E.H. Carr, to study thehistorian ‘before we read history’; that the documents, because of theirinherent textual limitations and our di"culties with authorial inten-tions, should be ‘scrutinised with care, taking into account . . . thenature of the document, and the context in which it was written’; andfinally, that historians do know that ‘perceptions and representatives ofreality are often very di!erent from, and sometimes just as historicallyimportant as, reality itself ’.51 Here again a moderate reconstructionist–contextualist is pointing out how deconstruction actually misrepresentsthe empiricist case. Quoting Joyce Appleby (in an echo of E.H. Carr),Stone asserts that ‘a text is merely a passive agent in the hands of itsauthor. It is human beings who play with words; words don’t play withthemselves’.52

Stone spoke for many historians when he said that his disagreementwith postmodernist history was when it claimed

that truth is unknowable, . . . that there is no reality out there whichis anything but a subjective creation of the historian; in other wordsthat it is language that creates meaning which in turn creates ourimage of the real. This destroys the di!erence between fact and fic-tion, and makes entirely nugatory the dirty and tedious archivalwork of the historian to dig ‘facts’ out of texts. It is only at this

What is wrong with deconstructionist history? 97

extreme point that historians have any need to express anxiety. Butsince nearly everyone . . . seems to be retreating from this position,there is now at last a common platform upon which we can all,without too much discomfort, take our stand.53

Stone here is making sense to most historians, few of whom insist onthe absolutism of the text. But to accept intertextuality, a historicistreading of the past, or an impositionalist and mediatory role for thehistorian, is not to wave a white flag in the face of the deconstructionistonslaught. A compromise with the deconstructionist critique is o!eredby the historian of France and leading ‘new empiricist’ Gabrielle M.Spiegel when she says that

If one of the major moves in post-structuralist thought has been todisplace the controlling metaphor of historical evidence from one ofreflection to one of mediation (that is, has been a shift from thenotion that texts and documents transparently reflect past realities,as positivism believed, to one in which the past is captured inthe mediated form preserved for us in language), then we need tothink carefully about how we understand mediation and how thatunderstanding a!ects our practice.54

If by mediation we mean acknowledging the historical and social cre-ation of a text as well as the need to evaluate it ‘as a literary artefactcomposed of language’ which demands ‘literary (formal) analysis’, thenreconstructionists and constructionists can both capture the reality ofthe past, accept the referentiality of language, and at the same time seetexts ‘as material embodiments of situated language-use’. In otherwords, historians can view texts as the material embodiments of thevarious uses of language, which reflects the ‘inseparability of materialand discursive practices and the need to preserve a sense of their mutual. . . interdependence in the production of meaning’. Spiegel calls thispractical realist compromise with deconstruction ‘the social logic ofthe text’.55

Less charitable to the deconstructive consciousness is John Tosh inhis support for the moderates in their attempt to compromise the twomainstream approaches. He maintains that significant advances in his-torical understanding are

more likely to be achieved when a historian puts forward aclearly formulated hypothesis which can be tested against the evi-dence. The answers may not correspond to the hypothesis whichmust then be discarded or modified, but merely to ask new questionshas the important e!ect of alerting historians to unfamiliar aspects

98 What is wrong with deconstructionist history?

of familiar problems and to unsuspected data in well-workedsources.56

All historical study is selective and, as Tosh would have it, ‘thereforepresupposes a hypothesis or theory, however incoherent it may be’.Because this process of hypothesis-making goes beyond the evidence, itis legitimate for historians to use ‘a flash of insight or an imaginativeleap, often the bolder the better’.57 None of this, of course, is anendorsement of the deconstructive emphasis on the cognitive func-tioning of narrative. For mainstream constructionist historians, thisCollingwood–Carr approach provides a more sympathetic definitionof history than positivism, and one which seems a reasoned andlegitimate riposte to the deconstructive stress on the linguistic turn.

HISTORY AS NARRATIVE

In the early 1970s, A.J.P. Taylor suggested that we historians ‘shouldnot be ashamed to admit that history is at bottom simply a form ofstory-telling. . . . There is no escaping the fact that the original task ofthe historian is to answer the child’s question: “What happenednext?” ’.58 Like E.H. Carr, Taylor believed that historians impose apattern on events in the shape of a dialogue. Between the events

and the historian there is a constant interplay. The historian tries toimpose on events some kind of rational pattern: how they happenedand why they happened. No historian starts with a blank mind as ajury is supposed to do. He does not go to documents or archives witha childlike innocence . . . and wait patiently until they dictate conclu-sions to him. Quite the contrary. His picture, his version of events isformed before he begins to write or even to research. . . . When ahistorian is working on his subject, the events or statistical data orwhatever he is using change under his hand all the time and his ideasabout these events change with them.59

Although he clearly accepted the impositionalism of the historian, Tay-lor did not directly address the issue of narrative structure as a form ofhistorical understanding. The closest he came to this was his claim thathistory ‘just like historical fiction is an exercise in creative imagination’.Taylor’s rider was, of course, that we historians are constrained ‘by thelimits of our knowledge’ and by the historian pushing himself ‘back-wards into time’ or empathising with the past. Nevertheless, Taylor wasastute enough to recognise, along with Karl Marx, that when historianswrite history, ‘our version, being set into words, is itself false’. He

What is wrong with deconstructionist history? 99

concluded: ‘We are trying to stop something that never stands still.Once written, our version too will move.’60 What Taylor was referring towas the perpetual revisionism that goes on in historical interpretation,but he was getting close to articulating the main inhibiting factor actingon historians attempting to understand the past – its organisation as anarrative which has the best ‘fit’ to the truth of what actually happened– what the historian Robert Berkhofer calls the or a Great Story.61

Taylor recognised that it is narrative that bridges the interpretative orcognitive gap between the historian and the evidence – a history and thepast. Pushing the argument further, do historians – mainstream ordeconstructionist – all simply o!er up a narrative of the past, or dis-cover its innate narrativised character? As Berkhofer says, can we getbeyond an imposed Great Story to the reality itself – the real story?

If we prioritise language before the content of history, the issue ofrelativism emerges with our choice of emplotment rather than withmatters ideological. In 1995, the American Historical Association pub-lished the updated third edition of its Guide to Historical Literature. Inits first section Richard T. Vann addressed theory and practice in con-temporary historical study, noting the recent greater consciousnessabout the role of narrative in the constitution of historical meaning.Vann recognised that the focus of interest in history ‘has largely shiftedfrom preoccupation with causation, explanation, determinism, andmoral judgements to the language historians use and the stories theytell’.62 He accepted that in the past thirty years the most influentialbook on historical method has been Hayden White’s Metahistory. Butin Vann’s view, White unfortunately presumed ‘to show that historicalevents could support any number of narratives – even narratives ofquite di!erent sorts’. Instead of dealing with empirical data ordered viasocial theory, history for White is created through poetic, emplotment,ideological and moral decisions. According to Vann, this ‘raised thespectre of relativism in a new way, forcing historians to confront theproblem of comparing possible narratives, each of which might becomposed entirely of true statements’. As a result, the new-found free-dom of narrative was bought at the cost of no longer being able ‘todiscredit alternate narratives by an appeal to the evidence’. This, inVann’s opinion, is not a price most historians are willing to pay,hence the antipathy with which much of White’s narrativistdeconstructionism has been received.63

While most historians agree with Collingwood that, although wecannot ever know historical truth, we are happy to accept ‘the obviousfact that we can and do substitute one narrative for another’, this isclearly not done on the deconstructive or relativist grounds o!ered by

100 What is wrong with deconstructionist history?

Hayden White. As Collingwood says, it is done ‘not on the grounds ofpersonal preference but on wholly objective grounds, grounds whosecogency anyone would have to admit if he looked into them, while yetfully aware that our own narrative is not the whole truth and is certainlyin some particulars untrue’.64 McCullagh agrees with Taylor (andCollingwood in this instance) that historians cannot get away fromlanguage and words:

almost all descriptions of the world use language [but this] . . . doesnot prevent their being true or false. In the case of literal descrip-tions, they are true if one of their possible sets of truth conditionscorresponds to what actually happened; and to say that they are trueis to assert that such a correspondence exists, that the world was as itis described.65

This position rejects the Whitean model of historical interpretationfounded on generic plot configurations, arguing that it is unlikely that,for example, the 1944 Warsaw Uprising could be interpreted equallywell as romance, farce or tragedy.66

This seems to be the position adopted by most mainstream his-torians. As the American historian David Carroll has said of the impactof Hayden White’s Metahistory:

it would be fair to say that the history profession as a whole hasrefused to take seriously any approach to history that has theappearance of being too ‘literary’ or rhetorical. Historians have forthe most part ignored or simply rejected the critical possibilitiesopened up by White’s . . . work . . . influenced by critical strategiesassociated with poststructuralist and deconstructionist theories ofdiscourse and textuality.67

But, having said that, Carroll agrees with the French historian PhilippeCarrard that more and more historians are (or they should be) aware ofthe literary or poetic as an important dimension or characteristic ofhistory-writing. The main lesson for mainstream historians today is tounderstand that ‘the goal of objective history is impossible to realise inlanguage’.68 Neither crude empiricism nor positivism are the routes tofollow in the attempt to overcome the rhetorical and figural characterof language and textuality. For Carroll, this recognition is not, however,a sell-out of history to the deconstructionists, but is a simple acknowl-edgement of history’s ‘epistemological and ideological assumptionsand limitations, on the one hand, and its formal, rhetorical operations,e!ects, and contradictions, on the other’69

This pro-narrative but anti-deconstructionist position is endorsed by

What is wrong with deconstructionist history? 101

moderate reconstructionists like Appleby, Hunt and Jacob who, withtheir ‘new theory of objectivity’, assume truth comes from the conflictof ideas ‘among diverse groups of truth-seekers’. Our pragmatic accessto this truth is through ‘the validity of each reconstruction’ whichdepends ‘upon the accuracy and completeness of the observations, not. . . perspective’.70 They argue that ‘To deny the writing of historyobjective validity because of the historian’s essential creative e!ort is toremain attached to a nineteenth-century understanding of the produc-tion of knowledge.’ While rejecting the postmodernist collapse of sub-ject and object and the deconstructive consciousnesses caricature ofhistory as little removed from nineteenth-century positivism, withMcCullagh and Carroll, Appleby, Hunt and Jacob fully accept the con-straints of language in the pursuit of truth. They recognise the poor fitbetween what happened in the past and the historian’s narrativereconstruction of it.

How does this poor fit look in practice? Historians, as Arthur Dantohas suggested, use narrative sentences to refer to events occurring overtime.71 Such a process requires using narrative to explain the causalconnections between events as the end product of the study of thesources and their interpretative contextualisation. For deconstruction-ists like White this process is flawed because the historian cannot cap-ture the past faithfully in language or as narrative, and the search fortruth is replaced by the truth-e!ects of the explanatory narrativedevices of figuration and style, emplotment, argument and ideologicalstatement. Another philosopher of history, Andrew P. Norman, rejectsWhite’s anti-narrative analysis of history, arguing instead that whilenarrative is the essential explanatory device for historians, its figurativenature does not mean that it cannot at the same time be literal, claimingthat ‘there is nothing contradictory in this’.72 In other words, languagein its narrative form is not a desperate illusion but is su"ciently in touchwith past reality to make the search for historical understanding viable.

Appleby, Hunt and Jacob accept this view, and that the historianconstantly makes literary choices in describing and evaluating the pastwhich ‘has a very strong influence on the way that evidence and argu-ments are presented’. As they say, form and technique are deliberatelyselected by the historian to implement arguments and make them con-vincing. Their text Telling the Truth About History was, they franklyadmit, written from a position intended to

go beyond the current negative or ironic judgements about history’srole. We as historians . . . [have made] our own aesthetic choices, justas others have chosen comedy, romance, or irony for their writings.

102 What is wrong with deconstructionist history?

We are emphasising the human need for self-understanding througha coherent narrative of the past and the need for admittedly partial,objective explanations of how the past has worked. In this sense, wehave renounced an ironic stance.73

In a rebuttal of Hayden White’s conception of the unavoidable literari-ness of history, these three appreciate that what they have writtenrequires ‘aesthetic or literary choices because they involve ways oforganising a narrative’, but they are sure that ‘history is more than abranch of letters to be judged only in terms of its literary merit’. Theyconstrue their history, written in part as a result of their literary choices,as being ‘political, social and epistemological’ rather than conformingto a classic literary emplotment type. In a summary of their moderatepractical realist reconstructionist position they conclude that theseliterary choices

are political and social because they reflect beliefs in a certain kindof community of historians and society of Americans. They areepistemological because they reflect positions on what can be knownand how it can be known. With diligence and good faith they mayalso be at moments reasonably, if partially, true accounts of the . . .past.74

As practical realist historians they accept that ‘social reality is culturallyconstructed and discursively construed in the first instance’ and that‘discursive or linguistic models throw into doubt the once absolutistforms of conventional historical explanation’, and in this fashion theyopen up the way ‘to new forms of historical investigation’, noting that‘Foucault’s own work is perhaps the best known example of such a newform with direct historical relevance’.75 Their attempt to meet thedeconstructionists halfway, to the extent of not ‘rejecting out of handeverything put forward by the postmodernists’, has its limits however.Postmodernism has not convinced them, as it has not most historians,of the validity of ‘linguistic determinism . . . the reduction of the socialand natural world to language and context to text’. They continue: ‘Ifhistorians give up the analogies of levels (the Annals school) or base-superstructure (Marxism), must they also give up social theory andcausal language altogether?’76 In other words, they do not accept thatdeconstructionism has cast genuine doubt on the power of narrative toexplain. This defence of realist narrative history has united both prac-tical realist reconstructionists and constructionists as they accept thenarrative impositionalism of the historian as an important dimensionof historical analysis.

What is wrong with deconstructionist history? 103

Many constructionists do still, of course, deny any explanatorypower to narrative, arguing instead that non-narrative history is theonly genuine history. Stanford, for example, maintains that narrative,defined as the description of events in their original or naturally occur-ring time sequence, cannot be analytical. Regardless of whether everynarrative (whether historical or fictional) requires heroes and the tra-cing of emplotted change over time, many constructionists remain farmore interested in mapping themes and structures that can quite legit-imately lack any sense, as Eric Hobsbawm said, of ‘directional ororiented change’.77 Most constructionists do, of course, have a sense ofchange over time, as well as having preferred ideas about the directionin which their description and evaluation of events takes them and theirreader (their history possesses a teleology). What this means is that theybelieve in empirically derived facts, and the need for a narrative descrip-tion of their meaning, while not accepting that such a description has acognitive power. As post-Marxist constructionist philosopher AlasdairMacIntyre has argued, it will not do to exempt stories from the criterionof truth: ‘It matters enormously that our histories be true.’78

So, while some Marxists may accept narrative as the vehicle for carry-ing historical analysis they would not agree that it provides the realmeaning of the past. Marxist Alex Callinicos accordingly rejectsWhite’s version of the role of narrative in historical analysis as anti-realist and ideologically loaded. Conceiving history as a fictive his-torical representation, where meaning ultimately derives from how it iswritten rather than according to the factual anchor of objectively dis-coverable and describable real events, suggests to Callinicos that Whitehas a sceptical and relativist (that is a postmodern) North Atlanticbourgeois liberal agenda! White is thus not equipped to tell fact fromfiction or, as Callinicos says, able to distance himself from ‘nationalisthistorical mythologies’, by which he means White’s treatment particu-larly of events like the Holocaust. In Callinicos’s view, as well as that ofother non-Marxist critics, White’s formalism and relativism make himincapable of distinguishing truth from interpretation, and fact fromfiction.79

Most constructionists, however, still tend towards the view of narra-tive held by R.G. Collingwood in The Idea of History. It is simply notgood enough, in his apt phrase, to ‘scissors-and-paste’ evidence to pro-duce, through a process of compilation, historical accounts.80 ForCollingwood, this ‘pre-scientific form of history’ is inadequate becauseit does not permit the historian to challenge the authority of thesources. In e!ect it creates historians who only practise a passive kindof inductivism (similar to the crude reconstructionist position). The

104 What is wrong with deconstructionist history?

genuine historical method is a process of question and answer, chal-lenge and interrogation through the application to the evidence of test-able theory. In other words, historical facts emerge from what Carrwould claim was an interpretative dialogue involving some form ofsocial/political categories of analysis. The evidence becomes a sourceof questions not answers, and history is an interrogation of the pastthrough the emplotment of an appropriate social theory. Facts are asmuch constructions as anything else in history. For constructionists,then, neither a conservative empiricism lacking in theory, nor a radicaldeconstructionism that denies social theory but relies on a flawednarrative, can prise open the reality of the past.

CONCLUSION

Moderates in both the constructionist and reconstructionist main-streams accept that language-use – whether writing the or a story of thepast – directly influences historical understanding. But this does notmean that history is just another kind of fictional literature. Acceptingthis would mean agreeing that history is epistemologically no di!erentto poetry, drama or television scripts. Practical realists reject the claimof deconstructionists that because a gap exists between interpretationand the facts, narrative must fill it as the constitutor of historical mean-ing. If historians continue to accept that their enterprise is about thestudy of the contextualised evidence and the valid use of social theorythat purports to explain the links between events, then to claim, asdeconstructionists do, that history as a text is analogous to fiction, isboth dishonest and bad logic. Thus the historian James A. Winn,following in the Collingwood–Carr tradition, maintains that decon-structionists and new historicists ‘tend to flog extremely dead horses’when they accuse historians of operating on assumptions that includebelieving that history is knowable, that words mirror reality and thathistorians insist on seeing the facts of history objectively. Few main-stream historians today work from these principles in pursuit of ‘theillusory Holy Grail of objective truth’ but strive only to ground ‘aninevitably subjective interpretation on the best collection of materialfacts we can gather’.81

What we might summarise then as the Collingwood–Carr-inspiredmoderate response to deconstructionism is a far more telling andreasoned response than that of Elton and Marwick, who continue tobluster about empiricism and its six principles. The deconstructive con-tention that primary sources cannot provide access to historical truthbecause of the essential unknowability of past reality is not at all

What is wrong with deconstructionist history? 105

convincing, given that all the available evidence demonstrates theopposite. As Appleby, Hunt and Jacob hold:

Assuming a tolerance for a degree of indeterminacy, scholars in thepractical realist camp are encouraged to get out of bed in the morn-ing and head for the archives, because there they can uncover evi-dence, touch lives long passed, and ‘see’ patterns in events thatotherwise might remain inexplicable.82

Historians today recognise that the act of describing an observationdoes not necessarily invalidate the truthfulness of that description.Equally, the historian is not a free agent, like a sculptor who can takethe clay of evidence and shape it however he/she likes. All this adds upto a historical relativism that has always been there, and historians whohave been living long before the deconstructionists moved into theneighbourhood and pointed it out as an unacceptable practice. It isnow necessary, in the interests of community well-being, to examinethe complaints of those deconstructionist newcomers about theirneighbours in greater detail.

106 What is wrong with deconstructionist history?

6 What is wrong withreconstructionist/constructionisthistory?

INTRODUCTION

In the last forty years historical explanation in a positivist mode hasbeen rejected in favour of that of narrative.1 I have suggested how, apartfrom the diehards, most reconstructionist and constructionist his-torians have become aware of this development. Although most main-streamers may still not yet accept that narrative is the peculiar form ofhistorical explanation – imposed stories – all accept narrative as thedominant form of historical reporting, while maintaining that the sixprinciples that form the bedrock of empiricist hermeneutics remainfundamental to the study of the past. Practical realists, while acknowl-edging the culturally provided nature of knowledge, still insist on thesanctity of the source (evidence) as o!ering an adequate correspon-dence to what actually happened in the past. While they accept thatfiguration exists in the representation of historical knowledge, theywill not deny the ultimate legitimacy of empiricist epistemology. Forhistorians of the deconstructive turn this is the flaw in their argument.History, rather than being a projection of the content of the past, is aprojection of its form. In this chapter I will address the implications ofthis for the established empiricist paradigm through the four headingsof history as a separate epistemology, historical evidence, the historianand social theory, and the significance of narrative to historicalexplanation.

EPISTEMOLOGY

Most historians today are at least aware of the doubts held by a numberof colleagues about history as a discrete empiricist discipline. Joan W.Scott, for example, in invoking the thinking of Michel Foucault, hasargued that by ‘history’ she means

not what happened, not what ‘truth’ there is ‘out there’ to be dis-covered and transmitted, but what we know about the past, what therules and conventions are that govern the production and acceptanceof the knowledge we designate as history.

She continues, ‘history is not purely referential but is rather constructedby historians’. With such issues as the linguistic turn and the explo-ration of gender in mind, she insists that history’s ‘standards of inclu-sion and exclusion, measures of importance, and rules of evaluation arenot objective criteria but politically produced conventions’. She directlychallenges the attempts by the ‘guardians of orthodoxy’ to maintain the‘unquestioned predominance for their point of view by insisting thatonly they represent “truth,” or “science,” or “objectivity,” or “tra-dition”, or “history-as-it-has-always-been-written” ’.2 In this fashionScott throws down the gauntlet to the supporters of the traditionalparadigm. Empiricism is less a convincing and timeless foundation forthe discovery of historical knowledge through its dependence on therelationship of the word and the world, than an increasingly threadbaredisguise for a rigid, exclusive and conservative ideological vision ofwhat constitutes history as a modernist epistemology.

The majority of historians still accept the six points of the empiricistcharter, while seeking to relegate history’s cognitive literary dimension.But not all philosophers of history accept the six empiricist tenets whilestill not accepting the deconstructionist position. Leon Goldstein, forexample, seriously doubts McCullagh’s justifications for empiricism,especially what he sees as its three fundamental assumptions: first, theworld exists independent of our beliefs about it; second, our percep-tions can provide an accurate impression of that reality; and third, thehistorian’s rules of inference and contextualisation are a reliable wayof arriving at new truths about reality. For Goldstein the first assump-tion is frankly meaningless, the second has no relevance to history at all,and the third does not exist in so far as there are no explicit rules ofhistorical inference that can produce proof positive. All the inferentialmethod can do is provide indications of a possible or, what may bemore accurate, plausible past based on the evidence.

Beyond Goldstein’s critique of empiricism the main thrust of decon-structionist history remains centred on the consequences of writinghistory. Hayden White, supported by other philosophers interested innarrative like Kellner, Rüsen, Carr and Ankersmit, has declared thathistorical explanation does not emerge naturally from the documents.History possesses no objective research method, the results of which arethen written up in a detached manner. In written history there is never

108 What is wrong with reconstructionist/constructionist history?

certainty of meaning. The meanings of history are to be found not, asKellner argues, in the traditional primary documents, but in the struc-tures of figurative representation. As Kellner insists, all history is ‘partof a story, an explicit or implicit narrative’,3 and it is up to the historianto emplot the events referenced in the documents. The issue of truth isto be found in the historian’s aesthetic ‘emplotment’ of the apparentlychaotic events of the past. To emplot means to turn a chronicle ofevents into a narrative whereby the events are explained and mademeaningful. The explanation and meaning emerges when the events areconstituted as one of the four archetypal emplotment forms. These fourcustomary emplotments allow the historian to explain ‘what happened’by transforming it into a story of a particular kind: an emplotment thatis romantic, tragic, satiric or comic. The ‘kind of story’ is, plainly, theemplotment selected. So, if a history is emplotted as a romance it is‘explained’ as a romance – and this becomes the truth of the story ofthe past.

Because, for White, stories are not ‘found’ in the evidence but pro-vided by the historian via emplotment, this issue of truth is crucial.White di!ers from the other key theorist of emplotment, Paul Ricoeur,who says the historian’s emplotment is a ‘creative imitation’ by meansof the plot of lived temporal experience. Ricoeur sees emplotment,then, as the imitation of past action. Both the leading literary narra-tivist theorists Gerard Genette and Seymour Chatman and the psych-ologist and narrativist Jerome Bruner would agree with Ricoeur when hesays ‘To make up a plot is . . . to make the intelligible spring from theaccidental, the universal from the singular, the necessary or the prob-able from the episodic’.4 This kind of narrative truth is not to be con-fused with statements of justified belief (factual statements), which donot exist in the same category as narrative truth. While history as con-ventionally understood requires truthful, factual statements, the issue isnever what are the facts (because, unless a historian tells lies, they aregenerally agreed) but how the facts are arranged – emplotted.

This means construing history as an aesthetic and poetic act ratherthan an empirical one, and it means accepting that writing history gen-erates a particular kind of historical truth rather than the truth. Mostmainstreamers still assume that the written medium is essentially trans-parent, and that it is time-wasting to explore narrative as a cognitivedevice. Like scientists, they prefer to draw attention to the reactionrather than the retort. It fell to philosopher of history Perez Zagorin tosummarise this iron law of empiricist historiography: ‘In history lan-guage is very largely subservient to the historian’s e!ort to convey in thefullest, clearest, and most sensitive way an understanding or knowledge

What is wrong with reconstructionist/constructionist history? 109

of something in the past.’5 It is in these terms that most historiansmaintain the fiction of history as non-fiction. Most prefer not to thinktoo deeply about the form in which they report their findings and evenless about the possible range of things they could say about (emplot)the evidence.

In responding to this it is worth reminding ourselves that the arch-empiricist Leopold Von Ranke’s emphasis on research into the sourcesdid not stop him recognising that research must result in an ‘acceptablestory’. As Ranke said:

History is distinguished from all other sciences in that it is also anart. History is a science in collecting, finding, penetrating; it is an artbecause it recreates and portrays that which it has found and recog-nised. Other sciences are satisfied simply with recording what hasbeen found; history requires the ability to recreate.6

Mainstream historians do not follow up this argument, that historymay be conceived of as both art and science, and do not remain alert tothe power of language to shape meaning and create understanding. Inspite of the power of language it is usually overlooked because theestablished paradigm characterises what historians do and believeaccording to criteria other than the aesthetic. As we know, in the twen-tieth century this has produced the view that history is epistemologic-ally concerned with finding out the truth and objectively confirming thevalidity of historical knowledge.

But this mainstream opinion has not gone wholly unchallenged. Ifnot in the terms used by Von Ranke, empiricist-inspired objectivity haslong been under intermittent attack from so-called relativists. In the1930s Collingwood and the American historians Charles Beard andCarl Becker argued that historical objectivity was a myth. As Colling-wood pointed out, history must have a purpose, and who else can dis-cover that purpose but the historian? For Collingwood, the historianuses evidence to isolate the intent behind actions, hence his/herempathic approach. Gathering pace since the 1970s, the new wave ofdeconstructionism – inspired in part by the postmodern surge andwhat Ankersmit describes as its narrativist philosophy – rather thantaking up these early Collingwoodian notions to challenge the recon-structionist consensus, has instead chosen to emphasise the structuresof narrative employed by the historian, which unavoidably implicateshim/her in what he/she creates. But the upshot for both Collingwoodand postmodern historians is that we cannot divorce ourselves fromwhat we observe. All historical interpretation is, therefore, pro-visional, relative and constructed. Deconstruction, as a historical

110 What is wrong with reconstructionist/constructionist history?

method, is the de-layering of these constructed meanings andinterpretations.

This peeling-back process seeks out that which is repressed in the text(primary or secondary) – not only what is hidden from the naïve readerbut also what is hidden from the intentions of author(s). The decon-structive historian seeks out that which is present in the text that runsagainst the grain of what, at first blush, it appears to assert. This self-conscious reflexivity seeks out that which is avoided and suppressed aswell as that which is openly de-legitimised and denied. We must con-stantly seek out that which, in the name of objectivity and rationality,the text is indi!erent to – what many historians call ‘the other’. Therationalist objectivity of Western culture in the twentieth century hasvisited death and destruction upon itself and other cultures on a hith-erto unimagined scale in the repression of ‘the other’ – Jews, Serbs,Croats, women, the poor, lesbians, immigrants, aboriginals, gays andmany other members of marginalised and persecuted groups. Inchallenging the six points of the empiricist charter, the deconstructiveconsciousness does not reject rationality or reason per se, but insteadsuggests that its exercise does not always result in rightness or will leadto truth. The deconstructive position does not reject historical realitybut questions our access to it, our apprehension of it and, therefore, itsmeaning. Deconstructionist history argues that there is always morethan a single truth. Finally, deconstructionist history does not declarethat there is no hierarchy of value, but declares instead that all arecapable of making di!erent and legitimate value judgements aboutwhat is right and wrong.

Commentators as various as F.R. Ankersmit, Peter Novick andDavid A. Hollinger have argued that the commonsense empiricistmodel for deriving historical knowledge, as founded on the belief inhistorical objectivity, has been substantially damaged.7 This is notbecause of an intentional conspiracy to attack history as a discipline,but is the result of the general postmodernist recognition that thenotion of scientific objectivity, as a measure of truth and a constitutorof knowledge, and which somehow exists outside social experience, isan assumption, and a doubtful one at that. Other philosophers of his-tory, apart from Leon Goldstein, have cast doubt on the nature ofempiricism as the basis for historical understanding. The Britishphilosopher of history Mark Bevir reiterated the telling point thatour accounts of our experiences rely as much upon our organisingcategories as on the experience itself. As Bevir said:

This does not mean that our categories determine what experiences

What is wrong with reconstructionist/constructionist history? 111

we have . . . but it does mean that our categories influence the way weexperience the sensations we have. We make sense of the sensationsobjects force on us using our categories. Because our experiencesembody theoretical assumptions, our experience cannot be pure, andthis means that our experiences cannot provide unvarnished data fordetermining the truth or falsity of our theories.8

I read Bevir as saying that empiricism is faulty as a method of acquiringknowledge because our understanding of that knowledge is alwaysinfluenced by our ‘theoretical assumptions’. From their di!erent tradi-tions, both Collingwood (how can we interpret the textual evidence soas to locate the intent behind human action?) and Derrida (how can weinterpret texts at all?) have contributed to the deconstructive conscious-ness which, as we have just noted, emphasises an awareness of theimportance of the impositionalism of the historian as well as theorchestration of narrative in the creation of historical knowledge. Afterthe millennium, under the impact of our postmodern condition, we areexperiencing a re-definition of the philosophy of knowledge and his-torical study because we are now facing foursquare the issue of themismatch between words and things. When historians say that they areconfronting the past, they are actually confronting language. Language,like memory, can recollect, but it can only be a substitute for reality.

On practical as well as formal epistemological grounds, the new cul-tural history accepts that change and continuity in the past may beexplained as a function of the discourse of the historian as much as bythe raw evidence or record of the realities of past everyday life. Theexplanations for cultural formation in the late nineteenth century inboth America and Europe are not only derived from the written experi-ence of political, religious or factory life, or of urban or rural livingconditions (real events under a description narrativised), or the dis-course of dominant and subordinate groups represented in the voicesof race, community, class and gender. This record and these voices areinterpreted by the historian through the structure of narrative he/sheselects and through which understanding is achieved. Historical inter-pretation means the past translated through narrative.9 The growingdialogue between history and literary criticism expands the naturalhorizon of the relationship between cultural change and our historicalknowledge of it. It also seriously questions history as an epistemologydistinct from its cultural practice and contamination by society’s needs,demands and power structures.

Our discussions so far have been framed as basic epistemologicalquestions about history as a form of knowledge. Do we really expect

112 What is wrong with reconstructionist/constructionist history?

historians to reconstruct the past as it actually was? Perhaps it is betterto view history as a kind of literature written in the name of seekingtruth? Can we, ultimately, believe in the past only because of the sub-stantial amount of agreement among historians about what happenedthrough the creation of historical facts? How do deconstructionist his-torians derive so-called historical facts, and what degree of reliabilitycan we place on them? Can history ever be objective?

EVIDENCE

In his book Child Loving: The Erotic Child and Victorian Culture, JamesR. Kincaid says that he is ‘less interested in reconstructing the past thanin examining what our methods of reconstruction might tell us aboutour own policies’ in the present.10 Annoying to hard-core reconstruc-tionists, Kincaid’s position is indicative of the deconstructive attitudetowards sources and method. Rather than accepting sources as relics ofpast truth, and empiricism as the only authoritative methodology cap-able of accessing their truths, Kincaid extends the horizons of the studyof the past by recognising its presentism. Such a mode of analysispermits the kind of sophisticated reading of sources undertaken byother historians like Carol Douglas Sparks. As a historian of the im-perialisation of Native American women, Sparks’ deconstructive con-sciousness reveals how her Anglo sources are racially skewed as imagesand signs of Indian women. As she says:

Deconstruction of these signs, or symbols of cultural significance,not only reveals the gender-coded fabric of nineteenth-centuryAmerican colonialism and its patriarchal environmental ethos,but also peels back layers of Orientalist imagery to uncover thehistorically ‘real’ women beneath.

Sparks qualified this insight by rejecting the popular version of Der-rida’s famous invocation that there are only texts not contexts, main-taining that ‘Textual analysis provides a useful tool in deconstructingsuch colonial imagery’, while reminding us that ‘this exegesis must befirmly rooted in a broader historical context which incorporates polit-ical, social, economic, and intellectual factors’. For Sparks, deconstruc-tion allows her to burrow beneath the ‘factual content’ of Anglo textslike letters, poems, memoirs, newspaper articles and even military andscientific reports to expose their ‘fictive origins’: the reality created bytheir colonial authors ‘drawn from their immediate surroundings, butfiltered through their experiences and expectations. Often, Anglo-American “fact” dramatically conflicted with the reality of Others’.11 In

What is wrong with reconstructionist/constructionist history? 113

other words, figuration once deconstructed can reveal much about thealternative and di!erent layers of historical meaning to be invoked bythe historian.

The work of Kincaid and Sparks alerts us to the misleading characterof the reconstructionist metaphor which likens our textual sources torock strata. The meaning of our sources cannot be chipped away atuntil we reach their real meaning. As another deconstructively con-scious historian, Roger Chartier, pointed out, texts do not conceal theirmeaning ‘like an ore its mineral’. Instead a text ‘is the product of areading and is a construction on the part of its reader’. The reader,whether the consumer of another historian’s writing or as the historianhim/herself, reading a source does not occupy the position of the text’sauthor by knowing his/her intentionality, but is likely to be inventing ameaning di!erent to that intended.12 Taking up McCullagh’s definitionof interpretation that we came across in Chapter 3, the interpretation ofa text involves an inevitable recombination of source, context (thetext existing intertextually with others), as well as the author’s inten-tionality. This can produce a multiplicity of legitimate meanings andinterpretations rather than necessarily lead to the true meaning.13

The sanctity of the sources is defended by the British philosopher ofhistory Mark Bevir, who declares: ‘good history depends solely onaccurate and reasonable evidence, not on adopting a particularmethod’.14 We are always thrown back on the evidence. It may well bethat Peter Burke’s deconstructionist suggestion has merit, that whileserving the available evidence it might be possible to make events inhistory more intelligible by following the novelistic method of tellingthe (hi)story from a multiplicity of viewpoints, rather than just that ofthe presumed omniscient historian – what he calls heteroglossia.15

Spiegel’s pessimistic conclusion, however, is that if texts can beexpected to reflect not reality but only other texts, ‘then historical studycan scarcely be distinguished from literary study, and the “past” dis-solves into literature’. Spiegel rejects what she takes to be decon-structionist extremism, preferring to retain a belief in a knowable pastreality while still acknowledging history as a written discourse. Spiegeldoes this by deploying the compromise of ‘mediation’.16 If past realitycannot be reflected (but is assumed to exist) then maybe it can be medi-ated, whereby texts do not transparently reflect the past but capture it in‘the mediated form preserved for us in language’. In other words,Spiegel accepts that texts, defined as extended discourses or culturalpractices, create meaning as between the real social world and our dis-cursive knowledge of it. The language of the text is the murky mediumthrough which we understand the past.

114 What is wrong with reconstructionist/constructionist history?

Spiegel eventually, but seemingly with a strong reluctance, acceptsthat language constructs the object rather than mediates or representsit. This is, of course, a reiteration of the common observation of thesocial construction of reality in, and through, language. As shedescribes it, rather than being extrinsic to both reality and interpret-ation, mediation in language is intrinsic to the existence and operationof the reality it creates, and all that this entails for the distribution anduse of power in society. Consequently, when researching sources we areactually studying already mediated discourses, composed of complexcodes of metaphoric meanings and understandings about how societyworks, intentions, and the role of history in it. We are not reading somepure reality of unclouded meaning – facts as linguistic fragments oftruth or historical reality – even less are we reading a text written out ofthe cultural flow.

Historians might, as a consequence, do well to conceive of evidence –texts and their intertextuality – as being ultimately determined by forcesthat are a highly complex mix of the cultural, the metaphoric and thenarrativist. To put it more simply, past cultural change may be mediatedthrough the written discourse(s) of not only historians, but also pasthistorical agents whose voices exist intertextually within their own lin-guistically mediated social, political and economic imagination andsituation. For all we know, it may be that the past as it actually happeneddid so according to a discoverable narrative emplotment which itselfmediates the principles on which knowledge and meaning are createdand policed. The interesting question then arises of whether we candiscover the dominant form of figuration and historical emplotment,which in turn will lead us to a fuller understanding of the arguments,rationalisations and ideological benchmarks held within our sources.

Historians, as Hayden White points out, have to use language as theyundertake this task, thus employing the same poor conductor of mean-ing as their sources. What this double bind does is remind the decon-structionist historian that we must not confuse either written history orevidence with the past. Historical interpretation ought not then attemptto recover the lost true meaning in the sources – the historian like adown-at-heel inebriate sorting through the debris of a dustbin in searchof the unbroken, and full, bottle. In accepting ‘the use of deconstructivestrategies in reading historical texts’, Spiegel allows that they are‘powerful tools of analysis in uncovering and dismantling the ways inwhich texts perform elaborate ideological mystifications’. A fullerunderstanding of the past might then be achieved by establishing ahistorical context ‘from other sources’. Saying this, in e!ect, reveals herundiminshed reconstructionist urge to infer the existence of the real

What is wrong with reconstructionist/constructionist history? 115

historical context within which she can deconstruct texts. Her rationalefor this is the provision of a complex strategy of research that willinterrogate text and context and thereby reach into the past for itsmeaning. In the end, the incipient reconstructionism in Spiegel’s argu-ment emerges as she confesses to accepting the past as ‘a once materialexistence’, although it is ‘now silenced’, while her embryonic decon-structive consciousness surfaces briefly with her claim that the past isnow ‘extant only as sign’ and one that draws to itself ‘chains of conflict-ing interpretations’ among those historians hovering over its relics.Spiegel thus tries to compromise the incommensurable. The decon-structive consciousness always doubts the correspondence between thesource and any presumed past, a position Spiegel always tries to avoid.

More convincing is the American historian David Harlan’s plainchallenge to the traditional understanding of what to do with evidence.Taking up a position derived from the German philosopher Hans-Georg Gadamer, Harlan maintains that historians can never strip theevidence of its accumulated meanings, nor by placing it in its contextexpect to rediscover its original author’s meaning. Evidence can never be‘severed from the interpretation through which it has been passed downto us’.17 It means realising that the historian is not able to place him/herself in an all-knowing, all-seeing situation, but rather contrives andframes, and is in turn controlled by, those discourses as power dispensa-tions, which predominate in any historical epoch or personal context.

This deconstructionist position is not unique in questioning the pro-cess whereby we ‘know’ reality past or present. It would be quite wrongto suggest that the deconstructionist approach to knowledge creation isthe lone voice speaking out against the correspondence theory of truth.Since the 1970s, so-called constructivists have challenged the contentand form issue in the generation of knowledge in both the physical andsocial sciences. Thus the ‘constructivists’ like the theorists of scienceBruno Latour and Steve Woolgar argued that scientific progress wasgenerated not simply through the mechanism of ‘discovery’ but were,rather, ‘constructed’.18 Indeed, led by the sociologist of science Latour,the radical constructivists argued that di!erent kinds of truth are cre-ated. The parallel with the creation of narrative truth (throughemplotment, which itself is generated as much by the ethical turn andideology as anything else) and factual truth is striking. Latour argues,for example, that narrative making is as essential to science as I amarguing it is to history creation. The position could not be much clearerin a ‘radical constructivist’ or deconstructionist age: the acquisitionof knowledge is as much a discursive and rhetorical activity as it isempirical. The argument that such ‘relativism’ must eventually destroy

116 What is wrong with reconstructionist/constructionist history?

itself because it is subject to the relativism it deploys is hardly con-vincing. If everything is relative, then that is just the nature of existenceand all claims to certainty are in the same situation – except the claim-ants for non-relativist truth (beyond the practically important but banal‘water freezes at a certain temperature’ claim) are in ‘relativist denial’.To be sure, we can possess experimentally and laboratory-tested data(actually, historians can’t, which makes it ‘worse’ for them), but we haveto do something with the data. So we end up in precisely the samesituation every time and the original meaning is always elusive. Whileothers have laboratories, all historians have is a text by which they try tocreate order out of disorder, meaning out of meaninglessness.

Reconstructionism cannot reproduce the original meaning, evenwhen the e!ort is made by historians like Gabrielle Spiegel to harnessit to a deconstructive awareness. Deconstructionist history is notreconstructionism looking over its shoulder. As Harlan points out:

if recent developments in literary criticism and the philosophy oflanguage have indeed undermined belief in a stable and determin-able past, denied the possibility of recovering authorial intention,and challenged the plausibility of historical representation, thencontextualist-minded historians should stop insisting that everyhistorian’s ‘first order of business’ must be to do what now seemsundoable. Historians should simply drop the question of whatcounts as legitimate history and accept the fact that, like every otherdiscipline in the humanities, they do not have, and are not likely tohave, a formalised, widely accepted set of research procedures, andthat nothing helpful or interesting is likely to come from attempts todefine one. If we ask, ‘what is historical writing?’ the answer can onlybe, ‘there is this kind of historical writing and that kind, and thenagain that kind’.19

Although talking about intellectual history, his words encompass allhistory, which

is concerned not with dead authors, but with living books, not with areturn of earlier writers to their historical contexts but with a read-ing of historical works in new and unexpected contexts, not withreconstructing the past but with providing the critical medium inwhich valuable works from the past might survive their past – mightsurvive their past in order to tell us about our present. For onlythrough such telling can we ever hope to see ourselves and ourhistory anew.20

Harlan thus o!ers what is the clearest of statements in defence of the

What is wrong with reconstructionist/constructionist history? 117

deconstructive approach to history. History is open to many ways ofstudying the past other than the belief that we can accurately reflect it.The study of history usually tells us as much about the historian’sconstructed narrative in the here and now as it does about past reality.

THEORIES OF HISTORY: CONSTRUCTING THE PAST

What does this deconstructive threat to the authenticity or purity of theevidence mean for the constructionist historian? It means a vigorousrebuttal. For hard-headed Marxist constructionist Alex Callinicos itnecessitates reiterating the well-worn argument that the historian satur-ated in the sources is well equipped to formulate hypotheses that can beverified through future siftings of the evidence. I have already noted hissuggestion that all history is theoretical. As he says, most mainstream-ers ‘draw (albeit in most cases tacitly) on theories about the nature andtransformation of human society’. He quotes Marxism, the Annalesschool and the New Economic History of the 1970s as examples of the‘self-conscious pursuit of a research programme in history’. What thismeans is that the historian deliberately, and promiscuously at times,deploys various kinds of social theory to ‘clarify issues that have arisenin her research, or even to define its objective’.21 This does not, heinsists, mean that history is reduced to hypothesis-testing by referenceto some general or covering law.

Callinicos, in his argument against deconstructionism, claims threethings: first, that it fails to acknowledge the materiality of history; sec-ond, that deconstructionism is flawed because it does not accept thatgenuine historical understanding relies on conceptualisation to explaincausality, and that neither tropes nor emplotment on their own canadequately explain anything; and finally, that deconstructionism fails togive due respect to the critical examination of the primary sources(Quellenkritik). The upshot is what Callinicos calls a ‘displacement’from the true nature of historical inquiry (that accepts a past reality towhich documents give access) to ‘the process of historical representa-tion itself ’, which is the extent to which our historical descriptions canever correspond to what actually happened in the past.

Deconstructionists in reply turn to other (non-Marxist) attempts toexplore the nature of constructionism, notably those of the philosopherPaul Ricoeur and historians Philippe Carrard and Robert Berkhofer.Ricoeur points out that constructionists still have to depend on narra-tive to explain the past, and that their analysis must eventually be con-figured as an emplotment.22 Carrard also notes how what he calls NewHistory (which I have designated here as constructionism) has signally

118 What is wrong with reconstructionist/constructionist history?

failed to eliminate narrative as the model with which to organise itsreports. As he says, ‘it is the plot that . . . provides a forceful answer toone of the central questions these [constructionist] texts are asking:How did we get where we are now?’23 The Annalistes, for example,have been quite unable to avoid the structuring e!ects of narrative intheir writing of the past. While they have conceptualised and re-conceptualised the past over several generations, they have been whollyunable to dismiss the power of language to e!ect meaning. Berkhoferaccepts that while so-called non-narrative histories transmit the cogni-tive e!ects of realism, by their aping of science they still cannot escapethe conventional devices of literature.24 Eventually even Callinicosadmits that facts ‘are themselves discursive constructs’.25

While defenders of constructionism attack deconstructionist histor-ians, it is not the advocates of the deconstructive turn who have createdthe doubts about the correspondence between the referent and its dis-cursive signifier. Such ambiguity would exist without the deconstructiveconsciousness pointing it out. Equally, narrative remains as the basicvehicle for historical understanding and explanation, and its tropes andemplotments will still be employed in writing history even though, asWhite says, it cannot give access to the truth. While it is possible toargue the constructionist position, that historical facts are arrived at byquestioning the evidence using an a priori theoretical construct, thisdoes not, as some Marxist constructionists claim, reduce history-writing to the level of mere narrative as if it were a secondary issue ofhistorical presentation. Deconstructionist history does not, however,criticise constructionism on the Elton ground that it pre-judges thepast, or is reductionist, or for that matter determinist, but rather arguesthat its results still have to be written down and understood asnarratives.

As we know, although Lawrence Stone misconceived the meaning ofnarrative in his 1979 article ‘The Revival of Narrative’, he did point tothe key non-Marxist trends in constructionist methodology then com-prising anthropological, ethnographical and structural history. Whenmoved to repeat the exercise in the early 1990s, he appeared to have aproblem not with the re-emergence (and explanatory failings) ofdescriptive story-telling, but with the growing authority of the decon-structive or linguistic turn itself. As he said, he parted company withthose historians ‘bedazzled by the lures of “discourse” [when theyextended their arguments] about the autonomy of “discourse” to thepoint of making it a historical factor in its own right’, thus allowing itto get in the way of explaining historical change according to the ‘morecomplex interactions of material conditions, culture, ideology and

What is wrong with reconstructionist/constructionist history? 119

power’.26 This is actually a rather good defence of the mainstreamconstructionist position. Stone went on to say:

As for the use of symbolic and social anthropology, influencedlargely by my friend Cli!ord Geertz, I can only repeat what I havesaid before. It has already had, and is continuing to have, a stunninge!ect upon historical scholarship.

Stone then catalogued the constructionist historians he had in mindlike Robert Darnton, Natalie Zemon Davis, Carlo Ginzburg,Emmanuel Le Roy Ladurie, etc. Next he quoted Simon Schama’s bookDead Certainties as an illustration of the blurring of ‘archival fact andpure fiction’.27 In e!ect, Stone overstated the case while failing toappreciate the complexity of the deconstructive consciousness whichnow increasingly inflects the work of the very historians he notes, par-ticularly Davis. It would be of interest to know Stone’s opinion ofSchama’s fascinating 1995 text Landscape and Memory which exploresthe relationship between landscape and history. What Schama calls the‘cultural psychology of nature’ may not be much to Stone’s liking as aform of history that is intensely personal and impositionalist incharacter.28

Since the 1970s, cultural historians using psychological, cultural andanthropological models of cultural analysis have blurred and blendedimperceptibly into the structuralist, post-structuralist and linguistic-inspired analysis of the poetics of culture using the textual metaphor.Influenced by Michel Foucault, what had hitherto been regarded ashard, objective social facts like race, gender and class (derived byQuellenkritik means) are now generally seen as culturally provided orsocially constituted. The constructionist notion of social theory provid-ing facts that reproduce the reality of historical life is now revised too!er access to the possible rather than the real nature of society. Thiscan be achieved at least equally as well by viewing society as a text inwhich events are presented as a complex series of discursive representa-tions, metaphors, symbols, icons, signs and rituals – all to be emplottedby the historian as cultural critic. It is not very much of an insight to saythat emplotment is part of a mix of pre-figuration, social theory,ideological positioning and empirical investigation. Facts as well asemplotment remain propositional for constructionist historians.

In his 1985 collection of articles Islands of History the cultural histor-ian Marshal Sahlins attempted to embed the idea of history as a textinto historical explanation by deploying a structuralist-inspired meth-odology which he called structural historical anthropology.29 The sig-nificance of Sahlins’ book lies in its attempt to wed constructionism to

120 What is wrong with reconstructionist/constructionist history?

a recognition of the relativism of history. Sahlins argued that history isculturally determined and that language plays a highly significant rolein that determination as well as its later interpretation. This text is aleading example of those that lead us to the conclusion that thelinguistic turn and the influence of post-structuralism have clearlymanifested themselves in the shape of the new cultural history.

The new cultural history, while being methodologically in debt tostructural anthropology, has also been liberated by Barthes, Derridaand Foucault, among others. For Foucault and Derrida the methods ofthe social sciences are quite inadequate, based as they are on an out-dated positivism. This is ‘logocentrism’, or the idea that there is a fixedmeaning (or explanation/cause) existing independently of language. Nomatter how complex and sophisticated are constructionist models ofpast social reality, the nature of the representation of their conclusionsis now contested, but not only by the deconstructionive consciousness.The constructionist philosopher Peter Burke has proposed that histor-ians examine the product of a marriage of narrative and construction-ism by

making a narrative thick enough to deal not only with the sequenceof events and the conscious intentions of the actors in these events,but also with the structures – institutions, modes of thought,and so on – whether these structures act as brake on events or asan accelerator.

He asks tantalisingly, ‘What would such a narrative be like?’30

HISTORY AS NARRATIVE

Burke believes that narrative can be a reliable vehicle for historicalanalysis if it can convey an understanding of wholesale changes insocial structures and institutions as well as single events. Burke insiststhat narrative can be a medium for social theory by pointing to novelistslike Leo Tolstoy and Shimizaki Toson. As he says, ‘it is likely thathistorians can learn something from the narrative techniques of suchnovelists . . . but not enough to solve all their literary problems’.31 His-torians cannot, as Burke points out, invent people, places or events andso must turn to ‘real’ people, places and events, but he concludes thathistorians will have to ‘develop their own “fictional techniques” fortheir “factual works” ’.32 Highlighting what he calls the micronarrativeproducing microhistory, which is the telling of the story of the lives ofordinary people, he cites Natalie Zemon Davis’ The Return of MartinGuerre as illustrative.33 In this instance Davis recounts the history of an

What is wrong with reconstructionist/constructionist history? 121

impostor who arrives in a sixteenth-century French village to claim themissing Martin Guerre’s life, wife and property. Davis deliberatelyemploys the micronarrative and fictional techniques to illustratebroader structural issues and the sweep of French peasant life. As shesays, the intention was to mix old and new cultural history. Such stories,like Geertz’s thick narratives, can thus illuminate big social structuralchange and continuity. The process of illumination is dependent, how-ever, on the power of the historian to use her subject-matter, argumentand, above all, her consciousness of the deconstructive power of figura-tive language. As Burke points out, historians in and since the 1990shave become ever more inventive in the way in which they employ thepowers of narrative to invest the past with life.

The vision of history as primarily an empirical and analyticallyfounded problem-solving epistemology is thus an ever diminishing one.The American philosopher of history Allan Megill points out in sup-port of Carrard that, even in the positivist world of the sciences,explanation commonly depends on the use of metaphor and descrip-tion, or ‘recounting’ as he prefers to call it. He defines recounting as theprovision of historical answers on the model of ‘telling a tale . . . a talefor the truth of which’ is attested to by evidence and argument. ForMegill, both recounting and explanation are to be found in the his-torian’s narrative mode of exposition. Megill and Carrard are makingthe same point as Lemon, Gallie, Mink, White, Ankersmit, and all theother narrativists, that while narrative is by its nature explanatory, thisdoes not mean that it is any more independent of the author or moretruthful than other kinds of explanation. We may make the point moretellingly by appropriating Simon Schama’s reference to the YosemiteNational Monument in the USA. As he says, we like to imagine thepark as empty of people even though the very act of painting or photo-graphing it ‘presupposes our presence, and along with us all the heavycultural backpacks that we lug with us on the trail’.34 By extension it isthe same with history. We may wish to imagine history as empty ofhistorians, a temporal wilderness, but every interpretation represents ahistorian backpacking through the past – the observer inside what he/she observes. Maybe we have reached the point where we are no longerexplorers but settlers?

While the idea of narrative as a report is generally accepted then, as amode of explanation I have suggested that it is far from universallyaccepted. The American historian and methodologist James A. Hen-retta maintains that many social historians remain ‘sceptical of theinterpretive range and power of the narrative mode of presentation’because of its ‘often impressionistic’ use of evidence, and its not being

122 What is wrong with reconstructionist/constructionist history?

‘amenable to quantitative or conceptual types of analysis’. Neverthe-less, even Henretta judges narratives to ‘embody a phenomenologicalperspective, and for that reason have been widely adopted by historiansworking in the pragmatic tradition’. He continues that ‘By placing asmuch (or more) emphasis upon the subjective perceptions of the actorsas upon the objective circumstances of existence, narratives underscorethe importance of human agency.’ He concludes that ‘historians whoadopt a chronological framework establish a basic congruence betweenthe lives of their subjects and those of their audience’. Henretta believesthat narrative is, therefore, important for the presentation of thefindings of research, ‘not primarily because of the absence of “jargon”but because its mode of cognition approximates the reality of every-day life’.35

In a footnote to his piece Henretta adds a significant gloss to thisprocess:

Although some degree of artifice enters into the construction of anarrative – in that the author already knows the outcome of thestory and therefore provides a false sense of open-endedness –artifice alone does not constitute a major objection.36

From the deconstructive perspective, of course, this is an issue rathermore important than Henretta suggests. The impositionalism of thehistorian may indeed make itself manifest as a matter of artifice inconstructing the narrative, but impositionalism operates at a muchdeeper level because of the narrative structure of history. As Ankersmitreminds us, the metaphorical (or tropological) nature of historicalunderstanding emerges from the historian’s ‘constitution of a linguisticobject’, what he calls the narrative substance and White designates asthe secondary referent. Historical interpretation is immanently meta-phoric because the historical narrative, while it is usually intended to belike the past, is, of course, nothing more than the historian’s surrogatefor the past. I am arguing that all historical narratives are representa-tions of cultural memories rather than mimes. Historical interpretationis nothing more than a re-presentation of those memories: an artifice,true, but one with a rather more substantive result than Henretta wouldhave us believe.37 This is a key deconstructive insight not to be dismissedin a footnote.

To extend this deconstructive argument we might note F.R. Ank-ersmit’s judgement that ‘History is no longer the reconstruction ofwhat has happened to us in the various phases of our lives, but a con-tinuous playing with the memory of this’. The point, as he says, is thatour ‘memory has priority over what is remembered’.38 That memory is

What is wrong with reconstructionist/constructionist history? 123

our written history. The inspiration behind mainstream history is anultimately unfulfilled craving. It is, as Ankersmit describes it, ‘the desireto discover a past reality and reconstruct it scientifically’, but these daysit is ‘no longer the historian’s unquestioned task’. Ankersmit concludesthat it is time for us to ‘think about the past, rather than investigate it’.This wish can be filled in substantial part by acknowledging the tropicand figurative frontiers in history writing rather than by gorging onempirical science alone.

In the production of historical knowledge, neither naïve reconstruc-tionist empiricism nor constructionist deductive or statistical prob-ability methodology can e!ace the impositionalism of the historian andthe constant issue of textualisation in the evidence of the past, or in ourrewriting of it as narrative history. It is impossible to avoid interveningin the past because of our translation of its traces into usable historicalfacts, akin to mixing colours and producing shapes on a canvas. Thisimpositionalism emerges not only as we compare, verify and con-textualise events, but also in subsequent narrative descriptions that pos-sess the reality – or truth-e!ects produced through emplotting. If thisvision is acknowledged, the disputes over the status and character ofhistorical study could, with goodwill on all sides, disappear. When thedefenders of the empirical paradigm accept history as a form of litera-ture – a narrative possessing ineluctably rhetorical, poetic and meta-phorical elements – they may come to recognise that this does notautomatically diminish history’s explanatory authority, or reduce theirprofessional status. The fact that figurative language may have nothingto do with past reality (because it is an allegorical or analogical descrip-tion of it) puts us in no worse a situation than does inference fromcontextualised evidence. Historians can still study a narrativised pastand seek to explain it. Many practical realists do now accept the poeticaspect of history – but as only one among many of the determiningfeatures influencing writing about the past. But, to place its poeticnature at the heart of the enterprise would, I suggest, strengthen historyas a discipline, rather than reduce it to some weaker project closer toliterature. Although most historians would still not accept the past as alived experience as being an essentially literary undertaking, whereby itscontent is understood primarily through its narrative form by people atthe time, as well as by historians later, the deconstructionist visionwould elevate form beyond the level of mere style.

As it was for Collingwood, and many historians of the deconstructiveturn today would still agree, the primary business of history is to studythe thinking ‘that goes on in the historian’s mind’, and we may wellfind ourselves accepting that what goes through the historian’s mind

124 What is wrong with reconstructionist/constructionist history?

constitutes the logic of the historical method.39 Essential to this processis the historian’s self-reflexivity in understanding not only the surfacereality-e!ects produced through style and figuration, but also thedeeper, and possibly determining, tropological structures identified byHayden White and Michel Foucault. What seems increasingly clear tomore and more historians today, as it did to Collingwood over half acentury ago, is that, on basic epistemological and linguistic grounds, theideal of the heroic model of science – the goal of objective history – isimpossible to achieve, and at best we can produce only an objectivitye!ect.40

Hayden White’s suggestion that historical method resides in theselection of a story of a particular kind rather than discovering thestory that faithfully reflects what actually happened paradoxicallyreinforces the strength of the argument of Paul Ricoeur and DavidCarr that we live in a narrativised culture. Although White remainsunconvinced by the argument of David Carr, that it is possible toobjectively discover the true story located in the sources, it seemsunexceptionable to argue that our everyday lived experience and historyare both saturated by narrative. This is not to repeat the modernistmistake of a foundationalism to which we can turn for objective cer-tainty, in this case by replacing empiricism with narrative. Simplyacknowledging the role of narrative in explaining our past and present-ing it to others is not a new kind of essentialism but an opening up ofthe past to new ways of describing it. Moreover, in the view of Lemonit is still possible to maintain objectivity rather than just project anobjectivity e!ect.

Lemon insists, opposing White, that it remains possible to explainwhat happened in the past objectively despite the inevitable process ofselection and emplotting of events, when the narrator-historian pro-vides enough information to make the whole process intelligible. Notunexpectedly leaping between events, so that they appear strangely orspuriously unconnected, constitutes for Lemon an objective account ofwhat happened. Such a connected narrative allows us to analyse andexplain beyond simple cause and e!ect because its structure permitsalternative interpretations. Saying that after President John F. Kennedywas provided with evidence of o!ensive Russian missiles in Cuba hethen established a naval blockade is rather less causally deterministicthan saying that because of the evidence of o!ensive missiles he set upthe blockade. This allows for the possibility of alternative policies beingopen to Kennedy. Lemon’s argument is that narrative explanation morefaithfully mediates the possibility of human choice or agency which heassumes is more realistic than the historian conjecturing a form of, in

What is wrong with reconstructionist/constructionist history? 125

this case, a strategic Cold War determinism. We may ask where doespostmodern history fit into this? After all, such a definition of narrativeis not too far removed from that of the traditional paradigm – narrativecan reveal empirical truths.

Deconstructively aware historians are at liberty to conclude severalthings about what we write and read as history. We could, as Lemon hasargued, regard it as the truthful/accurate representation of humanagency/choice in history. We could, like Hayden White, view historicalnarrative as just a defence for a bourgeois-inspired ideological prefer-ence for human agency. We could view the historical text as sharingwith the novel the important characteristic of authorial intrusion (ofthe imaginative historian), or as possessing no original author at all byassuming that how the narrative is framed depends upon the successivehistorians through whose hands (and minds) the text has passed. Wemay choose not to trust the language of our sources to correspond topast reality, and decide that we cannot reinsert our evidence into adiscoverable real past. We may seriously question the distinctionbetween fact and fiction in as much as both are the product of inter-pretative strategies and the emplotment of events. Although the fun-damental structure of factual and fictional narrative always remains thesame – as Lemon points out ‘this happened, then that’ – this will not tellus much about how the historian readily deals with the content. Lemonrepeats the question asked over a number of years by Hayden White:Can the narrative form in itself determine its content? Put di!erently,can the form of the historical narrative, however constituted by thehistorian, ever conform to the real narrative of the past as it wasactually experienced? Can we retell the story?

To suggest, as Hayden White does, that the past cannot ever be truth-fully accessed, because ‘the meaning of the story’ is directly influenced‘by the mode of emplotment chosen to make of the story told a story ofa particular kind ’, o!ends the strong empiricist sense of the historian’simpartial observer status in not pre-figuring the outcome of the story(or analysis as the empiricist would have it). For the inured empiricistthe imposition of a narrative structure through the choice of a particu-lar kind of plot structure – tragedy, romance, farce or whatever – brutal-ises the essential nature of the historical enterprise. Most mainstreamreconstructionists, as we are by now only too well aware, are concernedwith the correspondence between the event and their report, the sourceand an accurate corresponding statement about it. White’s further her-esy, that the choice of emplotment implies an unavoidable ideological orphilosophical commitment on the part of the historian, is the ultimateslur on the integrity of empiricists everywhere, given that historical

126 What is wrong with reconstructionist/constructionist history?

narratives should never be seen primarily as ideological forms. It isscandalous to those who believe in the reconstruction of the past as itactually was that writing history might be collapsed into ideology. Fordeconstructionist historians, on the other hand, the task is to explorethe lived narratives of the past, their tropic structures and explanatorystrategies which also include the nature of the social theory/lawsinvolved and the ideological implications that are likely to follow/precede.

CONCLUSION

The deconstructive consciousness, in accepting the non-referentialnature of language, unavoidably raises doubts about the traditionalreconstructionist/constructionist paradigms in several important ways.When language is viewed as constitutive of a meaning, rather thansomehow naturalistically reflecting real meaning, it follows that thehistorical narrative cannot generate fixed, absolute or truthful under-standing. In addition, although we recognise the impositionalism of thehistorian constantly and necessarily intervening in the past, we mayreasonably ask if it is possible to recover authorial intent in either anoriginal source or an interpretation. If the historical context cannotpromote the real meaning of the past, then we are forced to seek ameaning in the intertextual realm of the recoded and resignified evi-dence and the historical interpretation it generates. Both the source andits commentary remain ineluctably interpretative, and the traditional,empiricist-founded, reconstructionist/contextualist paradigm thatinsists on history as craft serves merely to disguise its poetic nature.Instead, the deconstructive consciousness accepts history as what itmight have been rather than what it actually was. When, as a historian, Iturn to the traces of the past, I cannot reclaim their real meaning – all Ihave is the tale I choose to bring forth from the sources which areimpregnated with the previous readings that I, and other historians,have of them.

History is first and foremost a literary enterprise. Its cognitive func-tion derives from the complex interpretative structure of narrativedefined as a set of proposals or suggestions about past events. Narrativeexplanation is quite unlike the constructionist version of historicalchange based on the belief in a functioning deterministic or causallaw(s). Constituting a narrative explanation requires the ordering, selec-tion and omission of events and occurrences, and by our study of howthe historian does this it ought to be possible to reveal something of his/her rationale or motive for producing this or that choice of narrative.

What is wrong with reconstructionist/constructionist history? 127

Deconstruction starts by revealing how traditional historians conflaterepresentation and referentiality. None of this precludes the decon-structively aware historian from believing that a past once existed.What it does mean is that he/she will write about a past within a self-conscious framework. It means accepting an impositionalism generatedthrough the historian’s dialogue with sources that do not necessarilycorrespond to the past and acknowledging that they are not projectionsof what actually happened because they are non-referential.

Narrative – the writing of the past – has at least one major dimensionremaining to be explored. It is to Michel Foucault and Hayden Whitethat I shall turn in the next two chapters to direct our attention both tothe rhetorical structure of history as the written form of the past and toone important implication that the past itself may be understood asconforming to the structure of narrative. If indeed the past is a narra-tive, from where does it derive, how does it work, and does the very ideafurther challenge history as an objective practice? Do we find the pastitself generated as a narrative by people in the course of their lives, or isit entirely imposed by the historian as he/she fabricates and orders his/her sources into a chosen form? These questions will be addressed in thenext two chapters.

128 What is wrong with reconstructionist/constructionist history?

7 Michel Foucault and history

INTRODUCTION

In this chapter I shall examine from the deconstructive perspectiveFoucault’s contribution to the study of the past, which has been toquestion the very nature of history as a distinctive epistemology byreplacing its empiricist-inspired inductive/deductive method withnarrative interpretation as the primary form of knowing and telling.Foucault’s anti-logocentric stance maintains that there is no unmedi-ated access for the human mind to a genuinely knowable original andtruthful reality.1 Our only door to experience (past, present or future)is through the primary medium of language as a signifying processnormally constituted within a framework for the exercise of power,legitimacy and illegitimacy. Derived from Nietzsche, this is a funda-mental shift from empiricism, because it entertains the impossibility ofknowing anything objectively, given that objectivity itself is a historicaland cultural construct.

Foucault is thus regarded by the majority of conservative as well asmainstream historians as anti-historical. This is not merely because ofhis refusal to privilege the modernist conception of scientific truth andtraditional categories of evidentially derived analysis, although hisrejection of the teleological and progressive assumptions of positivismcarves him out for particular suspicion;2 it is also because of his denialof linear historical causality between events and epochs (or epistemes ashe calls them), favouring instead a history based upon the disconti-nuities between dominant figurative structures operating in humanconsciousness. Such thinking is profoundly unappealing to the Anglo-American tradition. Foucault is further viewed with suspicion becauseof his doubts about the historian’s capacity to represent any version ofthe past accurately. The Nietzschean and post-structuralist lineage ofFoucault is revealed in his interest in what he sees as history’s dubious

quest for the origin of truth which is a part of the great myth ofWestern culture. Equally annoying for empiricists is his insistence –flowing from his historical methodology – that there cannot be anydistinction between what philosophers of history think and practi-tioners do. It is only when all history is self-reflexively engaged with itsown philosophy, and specifically the question of where our knowledgecomes from and how it is used (the framework of power), that we canconfront the questions he raises. Ultimately, as he says, the past con-strued as history is an endless process of interpretation by the historianas an act of imagination, and our categories of analysis, assumptions,models and figurative style all themselves become a part of the historywe are trying to unravel.

EPISTEMOLOGY

In his 1966 text, The Order of Things: An Archaeology of the HumanSciences, Foucault states what he believes to be the central epistemo-logical issue for history.3 He addresses how Western culture has organ-ised knowledge, and historical knowledge in particular. In assessing theimpact of discourse and social/cultural practice on the way in whichpeople in the past and historians order experience and memory, he asksthe inevitable structuralist question: What is the impact of language onhistory and experience when language is an arbitrary system of sociallyconstructed signifier/signified/sign relationships between the wordand world?

In practical terms this means that when Western thought firstacknowledged that such master concepts as ‘man’, ‘society’ and ‘cul-ture’ refer not to things (referents) but to linguistic constructs, then thehuman sciences all founded on reason, rationality, knowing, certaintyand inductive inference also became as White assumes, in e!ect, theprisoners of the historical figurative modes of discourse in which theywere composed.4 Foucault’s archaeological dig into the human sciences(especially the disciplines of medicine and history) lays open the figura-tive and narrative strategies that authorise their conceptualisations, toreveal what Hayden White calls the deep structure of their linguisticprotocols – the tropes. It is the historical succession of the tropes asinferred by White that constrain discursive practice, and condition thecharacter of each age (episteme) in terms of the creation and policingof knowledge. This indicates the endless character of the interpretativeprocess resulting from the situation whereby we can never scratch backfar enough to find the original truth. This, for Foucault, is the essenceof what we might now call the postmodern condition.

130 Michel Foucault and history

Specifically, in his studies of madness and medicine, Foucault exam-ined the ‘archive’ or evidence of the ensemble of discourses (verbal orwritten narrative statements) that constitute knowledge in any givenhistorical epoch (episteme). Foucault claims that historians shouldexamine the linguistic basis (i.e. narrative statements) that constitutehistory, rather than correspond to, or unproblematically represent, thereal world of things – that is, to abandon the search for original mean-ing. What could become an arid linguistic or narrative determinism is,however, modified by his acceptance of the culturally determined dis-cursive practices that provide the form in which our linguistically basedknowledge is produced. This shape or form is generated by the connec-tion between what historical actors say and do within the confines ofwhat society allows or rationalises to be true/false, right/wrong, legitim-ate/illegitimate. This is the idea of the social construction of reality, andalso what Foucault describes as the power/knowledge equation. So faras he is concerned, knowledge, comprised as disciplines, become con-trolling entities in our lives as they suppress and allow, exclude andinclude that which is not and that which is permissible. So, there cannotbe one history but there must be any number of histories of exclusion(the marginalised or ‘other’), inclusion (the accepted as normal) andtransgression (normal becoming abnormal).

The archive is taken by Foucault to be the body of narrativised evi-dence representing and signifying the episteme in which it was gener-ated, but which is, of course, encountered by historians within our ownhistorical epoch or episteme. Such source material cannot be inter-preted empirically in and for itself as an unproblematic point of origin.Historical evidence is to be understood not only for that to which itrefers (events as interpreted by historians), but as a vehicle by whichwe can grasp the deeper and more fundamental organisation of thelinguistic mechanisms underpinning the creation and constitution ofhistorical knowledge. History must recast itself as a literary and ideo-logically self-conscious process of thought. The epistemological signifi-cance of Foucault’s vision of history as a gateway to the human con-sciousness, as Gilles Deleuze claims, is not found in or through the workof the traditional historian because Foucault’s interest lies in seekingout the underlying structures, principles and ‘conditions governingeverything that has mental existence’, in this case ‘statements and thesystem of language’.5 Because the human consciousness works bymanipulating signs and metaphor, it follows that our understanding ofthe past works this way as well. This will never produce essentialtruths, only reveal the constant interplay of linguistic or narrativeinterpretation.

Michel Foucault and history 131

Foucault, like White, accepts the Nietzschean position (precursor topost-structuralism and deconstructionist history) that language as thefigurative power of the human consciousness is constitutive of bothhistory’s empirical content, as well as the concepts/categories used byhistorians to order and explain its data.6 Following on from this com-plex position of linguistic relativism (on the one hand) and modernistdeterminism (on the other), Foucault confronts each of the six pointsof the reconstructionist charter. Today, many historians who ac-knowledge the significance of the conventions of language-use and thenotion of history as a discourse – writing the past becomes the culturalpractice of history – accept (wittingly or unwittingly) Foucault’s argu-ment that historical interpretation is beyond the touching simplicitiesof the mainstream paradigms. Foucault confronts the empiricist charterby arguing that history is never objective because it cannot be inde-pendent of the historian and his/her own time or cultural context,and it is the power of language to create meaning rather than to dis-cover the true direction that history has taken that is important. As aresult, to be honest to him/herself and his/her reader, the historian mustavoid any claims to an empiricist-guaranteed disinterested objectivitylocated beyond the cultural frontier in which he/she lives.

The reasoning behind this position is Foucault’s sustained attack onthe reconstructionist belief in the adequate representation of realitythrough the narrative form. Not only is objectivity a myth, but moresignificantly we should recognise the sheer impossibility of the modern-ist theory of referentiality between word(s) and thing(s), statement(s)and evidence(s). In all this his main concern is to de-mythify history’sclaim to represent the reality of the past, and through it, its furtherassertion that explanation can in some way be complete, or reasonable,or realistic. As Michael Roth points out, this becomes clear when thosepossessing power make an appeal to history to rationalise their hold onpower.7 The legitimating authority of history is also used by those try-ing to gain power. Both the dominant and subordinate view history inthe same modernist fashion – claiming it as a rational statement oftruth – so as to deploy it for their own ideological ends.

Like Nietzsche, Foucault has come to accept that all modernist his-tory’s claims are ultimately spurious. In his important essay ‘Nietzsche,Genealogy, History’ published in 1971, he is particularly scornful of thee!orts of naïve empiricists to locate the historical truth which theybelieve to be ‘timeless and essential’. He argues instead that becausehistory is fabricated and we are implicated in it, we are wrong to con-clude that somehow we can stand outside history, or, what is worse, thatit is the essential requirement of our discipline.8 As he says, historians

132 Michel Foucault and history

take ‘unusual pains to erase the elements in their work which revealtheir grounding in a particular time and place’. He concurs withNietzsche that history should be ‘explicit in its perspective’ and shouldacknowledge that its ‘perception is slanted, being a deliberate appraisal,a"rmation, or negation’. This impositionalism of the historian is notmerely noted by Foucault, but celebrated. To be e!ective, the act ofwriting history must be interventionist and reconceived so that it mayopenly reach the ‘lingering and poisonous traces’ of the past ‘in orderto prescribe the best antidote’. It follows that this new history shouldnot be given over ‘to discrete e!acement before the objects it observes’and should not ‘submit itself to their processes [nor] does it seek laws,since it gives equal weight to its own sight and to its objects’.9 Thisvision of postmodern history not only rejects the fable of the cor-respondence theory, which maintains that the ‘truth’ is ‘out there’, butalso dismisses the reconstructionist belief in a transparent narrativethat permits the historical truth to emerge as if it existed beyond itsdescription. Hence Foucault dismisses the crude myths that flow fromthis general position: brute factualism, disinterested historians, objec-tivity, progress, stability, continuity, certainty, roots, and the demarca-tion between history, ideology, fiction and perspective. He rejects, in hisown words, empiricism’s will to truth.

In these terms Foucault attempts the de-thronement of both majortraditions of relating theory, interpretation and evidence around theaxis of the reality of the past. In typically robust language he puts‘proper’ historians in their place by claiming that as ‘the demagogue isobliged to invoke truth, laws of essences, and eternal necessity, thehistorian must invoke objectivity, the accuracy of facts, and the per-manence of the past’.10 So far as Foucault is concerned there is noundisputed content in the past, hence the need for greater self-consciousness among historians about the nature of our debates. In TheOrder of Things and The Archaeology of Knowledge11 his contributionto history is to provide histories based upon three basic and successiveconceptions of what he is trying to do. Initiallly he uses the designation‘archaeology’, then later he employs the term ‘genealogy’, and finally‘problematization’. The basic trademark of his historical method –which he calls his genealogy – describes how each unconnected anddiscontinuous historical epoch unilaterally imposes an intellectualorder on the generation and utilisation of knowledge.

History is no longer defined then by the established categories ofanalysis – economic structures, competing nationalisms, political andcultural revolutions, the march and opposition of ideas, great menand women, periods of excess and ages of equipoise, republics and

Michel Foucault and history 133

monarchies, empires and dynasties, famines and plagues – but insteadby how societies interpret, imagine, create, control, regulate and disposeof knowledge, especially through the claims of disciplines to truth,authority and certainty. Events do not dictate history: history dictatesevents. This radical conception translates into Foucault’s practical con-ception of an epistemic imposition on the past. This imposition pro-vides the intellectual culture in which society, ideology, technology andall human behaviour exists. The particular field that Foucault selectedin which to mount this challenge to traditional history is his study ofhow the treatment of the ill (a medical discourse) is related to its socialcontext (as it becomes a social practice), which he examined in his early1960s texts The Birth of the Clinic: An Archaeology of Medical Percep-tion and Madness and Civilization: A History of Insanity in the Age ofReason.12 So, how does Foucault’s conception of history work inpractice? How does he challenge the established and still dominantepistemological paradigm?

Foucault argues that people unconsciously organise and create know-ledge as discourses and practices within each of the four distinctivehistorical ages or epistemes that existed between the sixteenth and twen-tieth centuries. Each episteme was constituted out of the abstractionsof thought (concepts) that characterised the various disciplines, fieldsor branches of knowledge (by disciplines he meant the law, economics,biology, history, etc.) in Western thought. It is the function of hisgenealogy to unearth these epistemes and to locate the epistemologicalprinciples or concepts upon which the various fields or branches ofknowledge that compose them were and are built. Foucault notes threefundamental branches of knowledge – life (biological discourse),wealth creation (socio-economic discourse) and language (culturaldiscourse). The concepts that these branches of knowledge employ pro-vide the questions with which they interrogate their data and thus createknowledge.

Foucault argues that each of the disciplines is undergirded by sharedtrans-disciplinary mental attitudes towards the conditions of thoughtthrough which we organise all our knowledge. These orientations arecommonly referred to as our mental or intellectual senses of di!erence,resemblance and representation. It is this that leads him to the formula-tion of each episteme as an assemblage of concepts that fix and defineknowledge within its own epoch. These trans-disciplinary attitudes orconditions of thought are, of course, displayed in the pre-figurativetropes and narrative strategies we employ as historians, and whichcharacterise each epoch’s dominant form of narrative representation.Each age thus possesses its characteristic and dominant tropic

134 Michel Foucault and history

signature. All thought is contingent, therefore, on its origin in the tropicdeep structure of the mind. In this particular, Foucault is no di!erent toany other traditional historian trying to find the basis or origin to his-tory. The key di!erence is his insistence that epistemologically such atask of finding objective historical truth is going to be futile because ofthe collapse of the distinction between knower and known. As he will-ingly and I suspect somewhat pessimistically admitted, his own methodis, of course, subject to this collapse because it is modernist. But it is acollapse that the mainstream traditions deny. Unsurprisingly perhaps,the main feature of his own discourse on the status of history is foundin his reiteration of a key deconstructive principle with which we arenow familiar, the contention that the discourse of history exists within,rather than outside of, our culture and society.

The crude empiricist insistence that an accurate and accessible realityexists beyond the realm of interpretation is thus dismissed by Foucaultthrough his rejection of the belief that evidence corresponds with thetruth of what happened in the past. As we know, a true historical state-ment is defined by empiricists as a proposition that corresponds orconforms to the available and verifiable evidence, and in its turn thisdetermines the nature of the objectively written history text. Decon-structionist and other cultural historians prefer to explore the failure ofthe correspondence theory, especially in the relationship between evi-dence and context, knower and known – referred to as the intertextual-ity of written history – and a central problem to which I now turn.

EVIDENCE

In spite of his assault on the epistemology of traditional history, like allhistorians (including deconstructionist historians) Foucault accepts theneed to study the evidence in the archive. The essential proviso is thathistory’s facts are understood primarily as the epistemic discursive cre-ations both of people in the past and of the historian, written as therelationship the historian believes exists between words and things inany episteme he/she studies. This means that his/her understanding ofthe data results from, and can only be revealed in, his/her composed orinvented narrative which itself is ultimately a function of the tropicstructure of his/her own age. Because historical data are viewed then asrepresentations of events, not the events themselves, it follows then thatFoucault believes that historical meaning derives neither from theobjective historical contextualisation of evidence (discovering cor-respondences) nor from discovering the intentionality of the author(hence the death of the author).

Michel Foucault and history 135

Evidence, in the form of documents, is not to be seen as reconstruct-able traces of the past that are amenable to established inferential her-meneutics. History is the record not of what actually happened, but ofwhat historians tell us happened after they have organised the dataaccording to their own version of social reality. The reconstructionistdependence on the empiricist process of inductive inference – from theevidence to secure the truth of the past – is a counterfeit undertaking.The tropic determination of the episteme does not relegate theimportance of evidence, but inevitably places it in a secondary role tothe functioning of language – narrative form over content. Evidence,rather than being the point of departure, is history’s point of arrival.Metaphor is the point of departure.

Foucault views the evidence of the material world as the yield of thediscursive practice of the episteme. This means that evidence as eventsunder a description cannot generate brute facts conceived and under-stood as non-discursive relics, or bits of unmediated past reality.Foucault goes so far as to suggest that the concept of the empirical factis nothing more than a naïve discourse of nineteenth-century science.This means that how historians treat the evidence depends on the dom-inant linguistic protocol (or trope) of the epistemic archive they areworking with and within. So, while we must continue to study the avail-able evidence, it should be interpreted at its most fundamental level as amediation of the episteme’s conditioning narrative structures. It is theknowledge of those structures that forms the cognitive dimension ofthe deconstructionist historian’s linguistic turn, and informs the verynature of his/her historical enterprise. It is in this respect that I meanthat all history has within it an irreducible element of philosophicalself-reflexivity.

This recognition of the past as a written text also provides the plat-form upon which we may deconstruct the historian’s own explanatorynarrative. This point is now frequently made. Both the cultural criticAntony Easthope and social historian Patrick Joyce have flown the flagfor this argument. They contend (in Foucauldian fashion) that becausewe historians are in history just as much as anyone else, it is impossiblefor us to disentangle representation from content. As Joyce says, firstinterpreting then quoting Easthope:

For a fact to be accurate or not there does not have to be a relation ofcorrespondence . . . between discourse and the real. If the epistemo-logical debate is not resolvable, then there is no problem aboutdiscriminating accurate from inaccurate data, and tenable fromuntenable arguments. We do this all the time, widely di!erent

136 Michel Foucault and history

protocols obtaining in di!erent areas. None the less, these protocolsare themselves the product of history, logic turning out on inspectionto depend on ‘consensus and social construction (rhetoric)’.13

This once more raises the issue of socially constructed power relation-ships and their representation in language, the connection between thewill to truth and what Foucault calls the will to power.

Foucault argues that because evidence is always presented to us as theproduct of pre-packaged figurative codes, its character is dependent onhow people in the past and historians now choose to interpret an eventas either in conformity or conflict with accepted notions of humannature and/or cultural practice (sense of di!erence/similarity). In anygiven culture, at any given time, there are dispensations of power, bywhich some behaviour patterns are likely to be forbidden while othersare encouraged. These taboos or approvals, usually given a moral glossor rationale, are constructed for social and political purposes that haveto do with social power and its uses. As such, these cultural practicesdefined as thought and behaviour, either of which is forbidden orallowed, immoral and moral, are produced by an arbitrary system ofcultural encodation. This encodation, operating in narrative of course,is classified according to the polarities our metaphoric pre-figurationgenerates – our sense of di!erence giving meaning to objects under-stood as a continuity or as a contiguity (often viewed as a disconti-nuity), with an arbitrarily defined human nature as its measure. Aswe shall see in the next chapter, White agrees with Foucault that eachepisteme is almost certainly locked within a specific mode of discoursethat allows our access to the ‘reality’ in the evidence through ourencodation of sameness, similarity, resemblance or di!erence. It issuch encodations – representing ideological or social power con-siderations – that e!ectively create our sense of social reality.

It ought not now be too di"cult to see how short is the step from theinterpretation of evidence to the making of moral judgements, ideo-logical positioning and the functioning of power dispensations of dom-inance and subordinance in society. Defining historical interpretationas the recoding of the signifier and signified relationship – retropingthrough similarity or di!erence – prompted Foucault to instance thetreatment of the insane in Western European culture from the Renais-sance. This was one of the best examples he could find of the exercise ofsocial power and arbitrariness in cultural encodation. He dissected thefunctional nature and formation of such linguistic encodation thatresulted in the di!erent ways in which the insane have been viewed andtreated in each of the four epistemes. The point is that while historians

Michel Foucault and history 137

may care to arrange evidence chronologically, the interpretation of anypatterns found are not what actually happened, but are representationsof the figurative encoding process we have mentally captured and whichsurfaces in our narratives. No historian can escape the moral or ethicalconsequences of the arbitrary signifier–signified process. We are allimprisoned in the present as we narrate the past. This is the historian’sperennial double-bind. In Foucault’s terminology, to be e!ective, his-tory must acknowledge that it results from the perspective of presentlinguistically encoded power structures. It is this that gives history itsdissenting or, for that matter, status quo reinforcing nature in the hereand now.

This emphasis on our state of involuntary historical presentism per-mits us to understand how the encoding process is founded on thedispensations of power among dominant and subordinate groups. AsHayden White describes it, the writing of a historical narrative mustalways bear the imprint, in both its form and content, of the ‘shapinginfluences of language and cultural self-interest’ (my italics).14 What welearn, so far as our evidence is concerned, is that as historians we canrework its meanings only in the light of our immediate cultural experi-ences, and the ways in which that experience in part writes the past forus. The central lesson of Foucault’s genealogy is that we now have avision of history, a postmodernist history if you will, directed by therecognition of the cognitive authority of form, and that all ourattempts to obtain truthful representations are conditioned by linguisticand social perspectives. Consequently, no knowledge of the past can beobjective, and the world of the past cannot exist independently of ourrepresentation of it in the present.

In his texts Discipline and Punish and the first volume of History ofSexuality Foucault selects from the archive the discursive historicalconstruction of the social encodation of power over the human body –in prisons, mental institutions and sexual practices. Foucault reads theevidence for the changes it represents in the cultural attitudes towardsthe control of human beings and which mediated the epistemic andepistemological shifts in the construction of meaning in epistemic nar-rative forms. These two texts reveal much about Foucault’s historicalmethodology, in particular his incorporation of issues of power into theprocess of writing and constituting history. As a professionalised discip-line, history’s traditional function has been to organise the truthfulunderstanding of the past. However, as we undertake the act of writinghistory – whether as recovery, discovery, reconstruction, constructionor deconstruction – we are creating the discipline and the past. Foucaultacknowledges this by claiming that the historian’s discursive practices

138 Michel Foucault and history

constitute various subject positions that people in the past as well ashistorians may occupy.

Because each historical epoch or episteme is primarily defined bythe ways in which knowledge is generated and used – according to thefigurative process of distinguishing similarities and di!erences betweenobjects – language, ideology, power and the writing of history are,therefore, inextricably bound together in history. While mainstream his-torians study evidence in order to wring out its true meaning, to decon-struct means to seek out its multiple messages and through the exerciseof our imagination create possible ranges of meanings about referents.Hence, deconstructionist historians must always draw philosophicalattention to their narrative messages that possess all the ambiguitiesand possibilities of their evidential content and epistemic form.

THEORIES OF HISTORY: CONSTRUCTING THE PAST

From the outset of his career Foucault had ambitions to undercut thelogocentrist or Cartesian notion that knowledge derives from the way inwhich we, as human beings, take ourselves as the object as well as thefounding subject of that knowledge. His aim was to challenge our West-ern cultural reliance on the transcendental signified of the clinicalempirical method, the rules of inductive inference, the correspondencetheory of truth and reliable knowledge, historical objectivity, progressand disinterested analysis. As the Annaliste Roger Chartier has sug-gested after reading Foucault on the treatment of the insane, ‘Madness,medicine, and the state are not categories that can be conceptualised interms of universals’, rather they are categories of analysis, which arediscursive objects that are founded in the historical context – preciselyin the episteme.15 Foucault is thus o!ering both a speculative andanalytical philosophy of history that o!ers a particular process ofknowledge creation – the epistemic – as the foundation to our under-standing of historical change and meaning. It is the conception of theepisteme that is likely to have the greatest resonance for constructionisthistorians.

Despite his later strong anti-positivist, anti-empirical, anti-inferentialand historicist method, the young Foucault initially fell under the influ-ence of the Annales school. This stimulated what turned out to be hislife-long constructionist desire to discover the rules regulating collectivecultural practices. Upon reading Nietzsche, however, he turned fromspeculating on social theory explanations of the history of how peopleexperienced the material world, to studying the rhetorical world oflanguage in which they, as well as social theories themselves, existed. It

Michel Foucault and history 139

was his early constructionist-inspired focus that produced his specula-tive and analytical historical model of the episteme. In The Archaeologyof Knowledge Foucault o!ers the following definition of the episteme:

something like a world-view, a slice of history common to allbranches of knowledge, which imposes on each one the same normsand postulates, a general stage of reason, a certain structure ofthought that the men of a particular period cannot escape – a greatbody of legislation written once and for all by some anonymoushand.16

This great body of legislation constituting the principles or conceptscontrolled by this anonymous hand are elaborated in The Order ofThings, where he demonstrates how our lives conform to the conceptsinherent in what he calls the ‘empiricities’ or evidence located in thethree key human branches of knowledge noted above: life (or biologicaldiscourse), labour (socio-economic discourse) and language (culturaldiscourse). Given his later aversion to social science constructionism,Foucault is, in e!ect, claiming to have discovered in the evidence of thethree empiricities the four distinct epochs or epistemes, or stages ofreason, his ‘certain structure of thought’ that e!ectively constitute theepochal fields or branches of knowledge. We should be aware that thenotion of the episteme is not new with Foucault, having been previouslyelaborated by the Italian philosopher Giambattista Vico in the eight-eenth century. Foucault’s view of history has basically the same premisethat fired Vico – we can only be certain about that which Man hascreated. I shall return to Vico in the next section, but for the moment itis necessary to be aware of Foucault’s position that natural science mustalways remain ultimately unknown to us, whereas because we havecreated them, the human sciences a!ord our only chance of genuineknowing. Thus knowledge and history emerge from our own socialconstructions – in this instance Foucault’s notion of the epistemic/figurative basis of historical experience.

In his elaboration of the episteme, the point where he most radicallydeparts from traditional constructionist stage-theory history is his non-constructionist assumption that the four epistemes do not grow organ-ically out of each other, nor do they occur as revolutions in thoughtthrough some version of a dialectical process as they come into conflict.Instead they spontaneously appear in parallel to each other, filling inthe spaces suddenly vacated by other conditions of knowledge. In thisfashion we see an archipelago of branches of knowledge constitutingepistemes rather than a peninsula linked by bridges of causality – clash-ing classes, industrial revolutions, frontier experiences, catastrophic

140 Michel Foucault and history

famines, scientific discoveries, individuals bent on world domination,information revolutions, or whatever else most historians take to linkhistorical epochs. In the language of structuralism, Foucauldian historydoes not evolve diachronically, but is best understood synchronically, asan explosive discursive structure. While he notes that our access toreality through language is historicist and, therefore, doubts the possi-bility of a genuine knowledge about reality, even Foucault is forced stillto work from the reconstructionist assumption that the nature of his-tory, in his case the creative existence of the episteme, is indeed ‘outthere’. Foucault’s understanding of the four epistemes depends onknowing how language has developed and functioned in each to createand convey knowledge. Unfortunately, because we cannot escape ourown episteme, we can never know what constitutes historical change,for that is what happens catastrophically between epistemes. All we cando is map the past cataclysmic train of epistemes but, naturally, onlyfrom our own episteme – the postmodern – which regulates how weundertake that mapping exercise.

What is most unconvincing for historians is Foucault’s insistencethat the epistemes just happen with no apparent cause-and-e!ect mech-anism operating. This seems so counter-intuitive as to make little sense.But this worry is no need for concern. Given that we cannot locateepochal changes except through massive changes and dislocations inconsciousness, usually thought of as artistic, religious, scientific, ideo-logical or whatever, it is important to understand the nature of thatconsciousness as a linguistic mode of consciousness. According toWhite’s interpretation at least, Foucault is suggesting that human sci-ences (art, politics, science, etc.) are always captive of the figurativemodes of discourse in which they constitute the objects of everyday life(the four primary metaphors). So, when we seek to explain the changesbetween epistemes, we need to start with the dominant metaphoricuse of language. It is di"cult for historians to shake o! the belief,drummed into them from their first days at school, that change must beempirically determined. But this is not necessarily the case.

Disruptions in consciousness are what Foucault is concerned about.The epistemes do not succeed each other, the reason being that they arenot empiricities. New sciences of and attitudes towards the objects oflife do not emerge as reactions but as discontinuities. New sciences,values, attitudes, behaviours may appear empirically to rise up againsttheir predecessors simply because that is what empiricists look for. And,of course, they will find evidence of ‘reaction to’ prior beliefs and activ-ities because these undoubtedly exist. But how these are constituted isthrough the forms of expression that consciousness takes. Whether one

Michel Foucault and history 141

is convinced or not by this catastrophic theory of epistemic linguisticchange, the really important point is to address the fundamental role oflanguage in representing the objects of its gaze.

The first episteme, from the Middle Ages to the late sixteenth century(the Renaissance), characterises knowledge according to the dominantcultural/linguistic or narrative protocol of resemblance or similitude,where closely connected objects were viewed as part of what con-temporaries called the Great Chain of Being. As the range of com-parison (and the search for hidden links between objects) widened intoanalogy, a natural sympathy between objects was (arbitrarily) assumed– hence the notion of a linked chain. In the second episteme, from theseventeenth to the eighteenth centuries (the Classical), knowledge wasgenerated according to a linguistic protocol representing a clear senseof di!erentness. In this age, objects were understood and explained bypeople distinguishing them from each other so as to create meaningfulcomparison (like history discriminated from science); thus in this sec-ond age the formation of knowledge for Foucault is dominated bycontiguity and continuity. This second episteme is thus dedicated to thestability of classification and measurement, and particularly the ideathat order can be imposed on the real world primarily through thevehicle of a transparent language.

The third episteme, from the end of the eighteenth century to theearly twentieth (the Modern or Anthropological), did not evolve out ofthe Classical, any more than the Classical did out of the Renaissance.The epistemological breaks between epistemes, those unforeseen andcatastrophic switches in the grounding of knowledge, are witnessed inthe spontaneous emergence of the third episteme. Its preoccupationwas with Man as the central subject (and object) of reality. This pre-occupation is, for Foucault, best understood through the invention ofthe discipline of history through its typical modernist definition as theunderstanding of social change over time, as historians cast origins anddevelopment in the trope of di!erential succession. Foucault perceivesthe Modern episteme as creating a basic epistemological paradox forhumanity: Man as the product of his lived social experience, and alsothe constitutor of knowledge by the invocation of deductive knowledge.Such an epistemic tension cannot last for too long, and eventually theinvention (Man as both an empirical and deductive animal) will almostcertainly disappear as the idea of Man as a foundation of thought dimsinto obscurity, and knowledge (and consequently Man as a subject ofknowledge) is recognised as nothing more than an epistemic creation,and so we witness the deaths of certainty, history and Man as a know-ing animal. Needless to say, such apocalyptic visions do not impress

142 Michel Foucault and history

empiricists and which creates the impression that Foucault is a post-modernist, which he wasn’t. Indeed, his notion of successive epistemesis a very modernist idea.

The legacy of the Modern or Anthropological episteme in the inven-tion of the academic discipline of reconstructionist history is accom-panied by the naïve assumption of transparency in language and thebelief that narrative can objectively correspond with what actually hap-pened in the past. Taken together, these beliefs produced the predomi-nant nineteenth- and twentieth-century conception of history as anempiricist epistemology. From this standpoint, history, as understoodand practised by the mainstream, is but a vestigial remain of a previousconceptual epoch. Foucault’s provocation to this version of what his-tory is and does has been in part responsible for history’s present statuscrisis. It is this intellectual space in which this crisis of modernist historyexists that evidences the shift to the fourth episteme.

The present episteme, the fourth (may be the postmodern), if it hasnot already occurred, is in process of creation as the twenty-first cen-tury opens. Because Foucault is at great pains to insist that the epistemeis defined by the basal shifts that occur in the nature of language, anduse to which it is put, he argues that we come to understand history(defined as the mapping of catastrophic epistemic change) by examin-ing not its content, but rather the form or structure of the language inwhich that content is re-presented by people in the past and the his-torian. In his study of cultural change he noted the nature of powersituated in discourse but refused to trace the workings of power backto what mainstream historians would assume, in government, imperialcentres, or class struggles. Instead he sought them out in narrativisedemotions and instinct – specifically in the three empiricities of labour,life and language. From the study of these empiricities Foucault con-cludes that historical facts as well as constructionist theories (his ownincluded of course) can only exist as discursive entities, the products notof the process of inductive inference, nor less the rules of evidence, andeven less the correspondence theory, but as imposed linguistic or, to bemore precise, narrative processes.

HISTORY AS NARRATIVE

Foucault maintains that at the deep level of the human mind there is ahomology or parallel between the discursive construction of the threeempiricities and the epistemic organisation of knowledge. We know theworld in which we live only to the extent that we prefigure and narrateit to ourselves. History may, or may not, be merely a story we tell

Michel Foucault and history 143

ourselves for various social or power purposes, but equally it is possibleto conceive of it as a retelling of the emplotment of the lived past itself– history construed by people at the time through the dominant epis-temic figurative trope. Apart from confirmed anti-narrativists likeHayden White (who believes that there is no actual narrative in the pastto be discovered and retold), in reading Foucault we are reminded ofthe distant voices of philosophers of history like W.B. Gallie whoclaimed that it is the referent that turns a narrative into the story. As wenow know, Foucault’s version of history depends on it being under-stood as a language system of arbitrary socially constructed relation-ships between words and things, and through this process we create andlive out our own narratives. In his definition of the episteme, Foucaultplaces great significance on language in the constitution of knowledge(knowledge of our own lives and the world in which we live). In afurther elaboration of his definition he says:

By episteme, we mean, in fact, the total set of relations that unite, ata given period, the discursive practices that give rise to epistemo-logical figures, sciences, and possibly formalised systems; the way inwhich . . . each of these discursive formations . . . are situated andoperate.17

Language accordingly shapes di!erently the dominant modes of think-ing in di!erent epochs. Although we cannot ever know if thought isindeed the product of di!erent discursive or linguistic (tropic) forma-tions, it is possible to imagine that the central role in creating meaningmay be given to the discursive practices that constitute branches ofknowledge like history. However, as Foucault attempted to demon-strate, in each of the four epistemes the functioning of language isdi!erent. In the Modern Age the nature of language was questioned,with language becoming an object like any other, and with Foucaultaccepting the modernist conception that language cannot carry theweight of expectation that it will transparently represent the true orderof things or correspond to the past as it actually was. Language hasbecome just one more thing in a world of things. It possesses no insidetrack to reality. Its use by historians does not guarantee the accuraterepresentation of all other things. All it can o!er is the possibility ofemplotment (telling), and through it some kind of understanding(knowing). Foucault, like White, but unlike Gallie, cannot then tell thestory, only a story. It is this modernist denial of the representationalcharacter of language that forms part of postmodernism’s inevitableinclination towards uncertainty of knowing, and its desire to denyhistory’s traditional foundations.

144 Michel Foucault and history

This opacity of language not only makes it impossible to reconstructthe past as it actually was (or emplotted as it actually was?), but it mayalso account for the discontinuity between epistemes. This is becausethe branches of knowledge in each epoch are generated by using modesof representation based on di!erent narrative conceptions of the rela-tionship between the word and the world (our sense of di!erence). So,as imaginative historians of the linguistic or deconstructive turn, wemay examine the textualised evidence, and characterise the dominantform of narrative emplotment in each episteme. We may dig into pastepistemes to distinguish how events therein (cast as Foucault’s empiri-cities of life, labour and language) were explained by people to them-selves at the time through that age’s dominant and subordinatenarrative structures. We may then seek to understand how, in each epis-teme, the meanings invested by people at the time in life, labour andlanguage changed according to the ebb and flow of unconscious tropicforces operating beneath the level of their own explicit myth-making,empiricism or social theorising. The undertow of tropically inspiredepistemic rules that moved the thinking of people in the past – intel-lectuals, politicians, reformers, or whatever – would, of course, remainunknown to them. These tropic rules are, however, the fundamentalcodes that constitute the structure of narrative, the ‘total set of rela-tions that unite, at a given period, the discursive practices that give riseto epistemological figures, sciences, and possibly formalised systems’noted by Foucault.

At another point, Foucault claims that the episteme is constituted as‘the totality of relations that can be discovered, for a given period,between the sciences when one analyses them at the level of discursiveregularities’,18 emphasising the notion of the episteme as the mentalinfrastructure of all the human (non-scientific) branches of knowledge,the tropic historical a priori that can be revealed and constituted only innarrative representation. By definition then, because of its narrativeform, written history cannot avoid the use of the four primary figura-tive tropes: metaphor, metonymy, synecdoche and irony. Today manycritics, if not yet the majority of historians, believe that we have enteredinto an age (the fourth episteme) in which we cast the narrative of ourlives predominantly in the ironic trope – the dominant trope of thepostmodern episteme. Even some mainstream empiricists ask ironicallyhow is it that the social reality of the past can really be known to us – orindeed be represented accurately in narrative? This book, for example,could probably have only been written at this particular time. We knowthat Hayden White holds that relating text and context requires thehistorian to employ strategies of explanation that openly acknowledge

Michel Foucault and history 145

the defining tropes, emplotments and imposed formal arguments, all ofwhich carry, as Foucault has identified, moral/ideological/power impli-cations. If we follow Foucault, we are led to how the cultural signatureof each episteme works through the way in which similitude, di!erenceand comparison are characterised. Altogether this means that historicalexplanation employs the tropes, not just as stylistic figures of speech,but as pre-figurative strategies of explanation in their expression ofwhole–part, part–whole relationships. I am suggesting, therefore,that the troping processes of metonymy, synecdoche and ironyare the cultural signatures of pre-modern, modern and postmodernunderstanding.

Foucault’s linkage of epistemes with dominant tropes, however, isnot really a modernist (or postmodern) invention. Indeed, the basicidea of the episteme and the tropic foundation of knowledge originatedin the Renaissance episteme with the Neapolitan historian-philosopherGiambattista Vico. In his treatise The New Science (completed between1725 and 1744), Vico explored more fully than ever before the extent towhich language (as trope and narrative) represents things in the world,and also constitutes our understanding of the relationships presumedto exist between them. This notion was by definition lost in the Modernepisteme when the modernist myths of rationality and science weredecoupled from the cognitive power of language and rhetoric. The con-sequence, as White says, was to obscure ‘to science itself an awarenessof its own “poetic” nature’.19 Science, with modernist history aping itsmethods and sharing its mythology, assumed that it too could standoutside language and discover the truth of the past.

Vico rejected such Cartesian certainty and Enlightenment rationalityby emphasising the socially constituted nature of knowledge (Vicosubscribed to the opinion of verum ipsum factum, that the true and themanufactured are the same). The modernist corollary to the view thatscience is certain because it is man-made is science’s own absolute con-viction that truth must emerge from experimentation by Man in thephysical world with the aid of the calculus of mathematics. Because ofthe modernist preference for historical inductivism, and constructionistdeductivism, historians have failed, as Vico suggested at the time andFoucault subsequently accepted, to appreciate fully the fragility of thewritten form of the past and the ideologically bourgeois nature of theEnlightenment project. For Foucault, following what he understands tobe the logic of Vico, historians must be willing to suspend their belief inobjective proof in historical knowing, and instead accept a presentismand impositionalism that mediates and represents moral positions inthe world. Vico suggested that when historians write the past as a text

146 Michel Foucault and history

they unavoidably impose the present as its context. To try to avoid whatthey perceive as this particular problem, mainstream historians acceptthe Collingwood–Carr empathic approach to historical understanding.Deconstructionist historians, following Foucault, have instead found anopportunity here for further insights into the formative role of languagein the constitution of the past – rethinking the past unavoidably meansrethinking history.

Study of the four primary tropes provides an opportunity to familiar-ise ourselves with, as White puts it, the stages or cycles through whichconsciousness ‘passes in its e!orts to know the world’ but which ulti-mately always fails ‘to know it fully’.20 Such failure of knowing, whichFoucault also accepts, ought not stop us from writing the past. Weshould, however, write it reflexively, aware of the power of narrative toshape us ideologically as well as recognising that the past we are con-struing in narrative is not reality. As many commentators since Vicohave pointed out, but particularly Michel Foucault, language is theprimary vehicle for ideological dominance and opposition. Specifically,the deconstruction of the past hinges on our understanding of, as Alge-rian Marxist Louis Althusser describes it, the ideologically interpella-tive functioning of language – the capability of language operating atthe ideological level to place people in less powerful or subject posi-tions. Althusser maintains that the mass of people are constituted andlocated in situations of ideological subordinance because of the ideo-logical state apparatuses working through the media and other com-munication systems which are epistemically determined.21 Becauseknowledge is power, and the limit and the form of our knowledge isdetermined by the language we use to express that knowledge, how wedeploy language must give e!ect to what we think of as constitutingvalue, authority and legitimacy.

Put at its plainest, the tropes that prefigure (determine?) historicalwriting, as well as our understanding of the epistemic rules generatingknowledge, are ideologically saturated. For the historian who wishes tolocate the dominant trope, this means seeking out, in both evidenceand in his/her historical imagination, the relationship between form andcontent. We should understand that the reality of the past (presumed toexist) is a textually generated and ideologically tainted reality-e!ect. Bythis I mean that written history must grasp the nature and significanceof the troping process in the past as both a lived and written experience.Just as the tropes were the foundation of Vico’s stage theory of history(metaphor represents the age of gods, metonymy the age of heroes,synecdoche the age of men and irony the age of decadence and decline),so for Foucault there is a tropic foundation to each of the epistemes.

Michel Foucault and history 147

Not surprisingly we might view post-structuralism and deconstructionthemselves as postmodern empiricities, each in their own way rejectingman-centred modernist history, both evidencing the present status crisisin history.

CONCLUSION

In 1984 Mark Poster could claim, with some accuracy, that the mainreason for what he called the incoherence of historical writing is ‘theabsence of theoretical reflection by the practitioners of social history’.22

While this is no longer the case, by and large the mainstream are, how-ever, still recognisable by their wilful disregard of Foucault’s work.They deny Foucault’s insight that it is the role of the historian to locateand explore the archive’s discursive and non-discursive practices withinan episteme, so he/she may present to the reader the narrative trans-formations generated, and how they may have directly conditionedevents, actions and beliefs in the past. While Foucault may or may notbe the first deconstructive or postmodern historian, he is qualified tobear the title if only because he points to the epistemological breakbetween the Modernist Age and our own. It may be that the title ismore deserved because of his refashioning of history as a form thatdoes not rely on the inductive inference or imputation of causality,origins, certainty and truth. Consequent upon this rejection of empiri-cist foundationalism we can rethink the nature and purpose of his-torical evidence as an archive of discursive and non-discursive practice,recognising that its utility resides in what it ultimately tells us aboutthe organisation of knowledge according to criteria other than thecorrespondence theory of knowledge.

This new view of the nature of history seemingly flings philosophyinto the face of practitioners: as Elton said, bringing it out of thescullery into the drawing room. Historians after Foucault are increas-ingly coming to terms with their study of the past, and the past itself, ascomposed narratives. It is in line with this new consciousness that Hay-den White has established his own contribution to the writing of his-tory. His assertion, following Foucault, is that while narrative as a formof cognition and representation is unavoidable, it is also infuriatinglyrecalcitrant. However, if history as a discipline is not to be confusedwith the past – a central plank of White’s programme – then its possiblerealities can be accessed only through the conceptual powers of his-torians as constrained by linguistic structures and categories. It is toWhite’s narrative model of historical explanation that we now turn.

148 Michel Foucault and history

8 Hayden White anddeconstructionist history

INTRODUCTION

Historians who normally reject the idea that the form in which theirresearch is written up creates historical meaning, do so on the assump-tion that the language used to write about the past can correspond tothe past as a narrative. This is a view rejected by Hayden White (alongwith others like Louis Mink, F.R. Ankersmit and Paul Ricoeur).White’s analysis of how historians, as they describe and evaluate pastevents, e!ectively invent the past is probably the most radical develop-ment in historical methodology in the last thirty years. It has forcedother philosophers and historians to address the issue of the cor-respondence or homology between narrative form and lived experience.For White, because the past is invented or imagined rather than found,history the first time around does not conform or correspond to a pre-existing narrative or story. White does not dispute that the past existed,and he is not anti-referentialist, but his answer to the question I posedat the outset, asking whether the past pre-exists as a story told bypeople in the past to explain their lives to themselves, is to argue that weimpose stories on the past for a variety of reasons which are explana-tory, ideological and political. Narratives are not detached vehicles fortransmitting past realities, nor less can historians discover the truenarrative of the past in the evidence of human intentions and beliefs.

Why is it important to consider White’s arguments about the natureof history? Well, he was the first to construct a detailed theory of his-tory as a tropic exercise. To understand the nature of history to itsfullest, we have to be aware that what reconstructionist and con-structionist historians claim are the central planks and tenets of (theirview of) history are only elements within it. As White has pointed outmore clearly than any other historian in the past half century with thepossible exception of William Gallie, because it is a narrative-making

exercise, there is far more to history than empiricism and inference.Reconstructionists are particularly upset with White’s argument thathistory cannot correspond to a given or pre-existing story of the past,much less one that is knowable for what it really means. For White,there is no meaning in the past. The historian provides this. What issignificant in this is the historian’s own existence. White is here acknowl-edging the disintegration of the distinction between subject and objectin much the same way he accepts the collapse of the di!erence betweencontent and story. Deconstructionist historians, like White, do not denythe necessity to reference the past through the available evidence (unlessthey are deliberately experimenting, of course) or that meaning andexplanation cannot be supplied for it. Arguably, only reconstructionisthistorians would categorically deny that meaning is provided for thepast. It is for this reason that the provision of explanation and meaningdoes not follow the empiricist logic of discovery but, rather, the logic ofnarrative making.

Historians within the two main tendencies believe in recoverableauthor-intentionality, truth, causes, origins, an adequate correspon-dence between words and the world, and insist that history emergesfrom the ultimate freedom of people in the past to act, think and makerational choices (or explicable irrational ones) not absolutely con-strained by material conditions like class. From this ideologically con-servative position they insist that there is a genuine, empiricist-foundedhistory methodology. However, some seem to want it both ways byinsisting on this set of beliefs about method, yet nevertheless wishing toclaim that ‘there is nothing much to say about historical explanation;nothing that cannot be said about explanation in everyday life’.1 Thisinconsistency ignores the twentieth-century debate whereby analyticalphilosophers have argued over the proposition that language is the pri-mary state in which knowledge is produced and understood, and thatthe structure of the historical text as a whole, rather than just its micro-level of individual statement about intentions, may be cognitive, that is,create meaning.

We may, of course, choose to challenge White and ask again to whatextent is the historical narrative actually homologous to the past – canhistory recover the story of the past, or do we merely impose a story?This would be a di!erent correspondence to that imagined by empiri-cists – what I will call a narrative correspondence – the extent to whichour story matches the story of the past not via empiricism, but via astudy of past dominant/subordinate rhetorical structures. The recoveryor match is, in the way of things, itself subject to the structure ofnarrative that permeates written history today. In this chapter I will

150 Hayden White and deconstructionist history

address this issue and White’s rejection of the idea of the historian’sdiscovery of the story. As F.R. Ankersmit has suggested, not until rela-tively recently has the historical text ‘as a whole’ ever been ‘the topic ofphilosophical investigation’.2 This is regrettable, given that history is theliterary activity most suited to this level of textual analysis. It is at thislevel that we address what is perhaps the central question for White andall those interested in the role of narrative in history: does life itself havea narrative structure and is it recoverable? If it does, and is, then historyas it is written must not be viewed as either the report of an objectiveempiricist research programme or a subjective piece of literature, but asa representation of past life and culture.

EPISTEMOLOGY

The recalcitrant nature of the signifier–signified relationship, the unfix-ing of referentiality and empiricist correspondence, the decline in cover-ing law constructionism, and Foucault’s exploration of the relationshipbetween the past and narrative form, have prompted the examination ofthe complex ideological, explicatory and emplotted structure of historyas a form of explanation. So it is that we view history as a literaryartifact rather than as the unalloyed product of contextualised sources,or a reconstruction of the empirically derived and accurate experience,or the construction of a social theory. Like Foucault, Hayden Whiteplaces his emphasis in the writing of history upon discursive practicesand determining tropes, o!ering a formal model which, when takenwith Foucault’s vision, allows historians to relate the structures ofnarrative representation to the nature of historical change.

In epistemological terms, White’s engagement with Foucault’s notionof the episteme, together with his own formal narrative model oftropically engineered historical explanation, raise important questionsabout the past as a textual product. White’s model of historical writingand understanding is now well known.3 Briefly, White o!ers a model ofhistorical narrative in which its form is taken to prefigure the historian’sunderstanding of the meaning of the content of the past. White’s keytext is Metahistory, published in the early 1970s. In it he demonstratedhow a historical narrative endows itself and the past with meaning. Formost historians a clear literary style in their historical explanation istaken as a measure of narrative’s irrelevance to understanding. When itcomes down to it, why study the lens rather than the object of studybeyond it?

White, however, forces us to confront the fundamental issues. Doeslanguage act in opposition to our assumption of reality because it is

Hayden White and deconstructionist history 151

only through language that we can apprehend that reality? (Yet,because of its figurative character, it must always fabricate that reality.)Does the prison house of language mean that we can never escape totruth? Does the form of our historical reconstruction directly conditionor constitute our interpretation? How do we impose our own narrativestructures on the past? White has not been alone in considering thesequestions. As a hermeneutic issue the functioning of narrative has beenaddressed by many historians. Frederick A. Olafson, for example, haspointed out not just the significance of the historian’s understanding ofthe language of the primary sources, but its importance in framing andanswering our questions. Through language we historians are central tothe process of creating historical understanding.4 White takes this up inarguing against the maintaining of a language–world distinction. Whileit is certainly important for historians within the mainstream tendenciesto claim adequacy in the representation of the past, White disputes thisempiricist investment directly.

Reconstructionist and constructionist historians are much troubledby what they see as White’s cavalier attitude to truth and the reality ofthe past. Of course, White is greatly concerned with both, but hisresponse to the correspondence theory of truth is what dismays mosthistorians. White stresses, in defence of narrative as a legitimate modeof representation, the communicative or representational function oflanguage. He suggests that the demand for the de-rhetoricisation ofhistory is essential to the realist as the only means by which subjectivityand political bias can be eliminated from history. By this mechanism,history and fiction can be readily distinguished, which is the first steptowards objective history. The realist assumption is that poetry has tobe excluded from history writing. In this realm of history language, thematter of bias does not exist because the stories that historians tell areall found in the evidence.5 Consequently, the issue of the historian’slanguage could not arise. It was simply the result of ‘rules of evidence’.However, as White has argued, given the nature of language, historianscannot escape from it into the ‘rules of evidence’. The notion of ‘imagin-ation’ in a poetic sense is thus supposedly removed from the process.

As I have already indicated, White’s historical method works fromthe general assumption that written history is unarguably a literaryenterprise and we cannot gain access to what the past was about otherthan through it. It follows that we understand the past through thenarrative form we devise to organise it. History at every level, therefore,is a text possessing an imposed or invented meaning. The decon-structive historian’s function remains that of interpretation, but aninterpretation viewed as the translation or rendition of one text (the

152 Hayden White and deconstructionist history

past) into a new narrative version which is another text of the histor-ian’s own invention (written history). This textual rendering of the pastis, as we are aware, steered by the four master tropes of resignification –metaphor, metonymy, synecdoche and/or irony. Because there is nonecessary correspondence between words and things, or language andpast reality, the historical text can be linked only to other historicaltexts and derive its meaning from those other texts. This is the argumentthat history gains its meaning intertextually, and while it reveals thefull import of Derrida’s play on the concept of di!érance, it alsoemphasises the insight of Foucault that the past may itself be regardedas a text. Of course, none of this stops the reconstructionist in his/hertracks – nor should it. As White points out, history defined as a verbalmodel can still be ‘o!ered by the historian as a representation andexplanation of “what really happened” in the past’ if that is his/her wish.Neither White nor the deconstructionive consciousness is embargoingthe evaluation of the past’s content as such; they are only questioningthe empiricist claims to an ultimate referentially and knowing history.

It is White’s contention that to understand what the past was aboutwe must impose a narrative upon it; hence our knowledge of the past isthrough a poetic act. This is the element of fiction in all historicalaccounts, and it is this part that is so abused through the neglect ofhistorians. Chosen by the historian, and constituting the fictionaldimension to historical understanding, narrative o!ers a subtle render-ing of pastness. The fictional element emerges both when empiricistsclaim to be representing the story as it actually happened and when thedeconstructionist choice of emplotment is taken to represent a story ofpastness. It follows, if White is correct, and people in the past do notactually live stories (that is, they do not impose emplotments of a par-ticular kind on their lives and times in order to make sense of them),that the reconstructionist argument that they have discovered the realityof the past in their story is undermined in as much as there is no story inthe past to be discovered. White insists that the past as history is not thestory – it is the fictional invention of historians as we try to recountwhat the past was about. As he says:

Historical situations are not inherently tragic, comic, or romantic.They may all be inherently ironic, but they need not be emplottedthat way. All the historian needs to do to transform a tragic into acomic situation is to shift his point of view or change the scope of hisperceptions. Anyway, we only think of situations as tragic or comicbecause these concepts are part of our generally cultural and specif-ically literary heritage. How a given historical situation is to be

Hayden White and deconstructionist history 153

configured depends on the historian’s subtlety in matching up a spe-cific plot structure with the set of historical events that he wishes toendow with a meaning of a particular kind.6

At another point White is adamant that ‘No one lives a story’ and thusall emplotments are imposed later by historians.7 We do not, of course,have to agree with White. History viewed as an essentially literaryendeavour may not preclude the possibility that people living in the pastdid indeed explain their lives to themselves as narratives as construedwithin their particular episteme. It follows that there may be some kindof narrative correspondence possible between past events as lived andtheir history as emplotted later by historians, but it could also meanthere are a variety of stories, or possible emplotments, in the past thatare generally constrained by the deep narrative structure of the epis-teme. Indeed, as White asks, is there an analogy between ‘the dynamicsof metaphorical transformations in language and the transformationsof both consciousness and society’?8 This suggests that stages in historymay well coincide, in some more or less determinist relationship, withtheir ascendant figurative tropes, with the variations from metaphoricto metonymic, and synecdochic to ironic expressions of thought associ-ated with the fundamental cultural transitions in Western society sincethe Renaissance. Are there, in e!ect, emplotments to be recovered fromthe past?

If indeed emplotment is tropically determined, and if Foucault iscorrect in his analysis of epistemic historical change, then White is in abind. As Robert Berkhofer points out, if White’s flow of tropes (follow-ing Foucault’s sequence of epistemes) do forecast or prefigure the aes-thetic and level of emplotment, the cognitive level of argument, andtheir joint ideological implications in historical accounts, then pastevents can only be construed as poetic acts independent of the realcontent of the past.9 That the tropic process directs the historian’s ownfigurative understanding of the past and people in the past may be seenin recent applications of White’s model to America’s modern historicaldevelopment.10 The use of White’s rhetorical constructionism does notmean that we cannot study what the past was about, but means that wemust recognise the severe limits of empiricism. Of course, as we decon-struct the past, we can deconstruct each other’s rendering of it. One ofthe radical benefits of deconstructionist history is its breaking down ofthe barriers between its own form and the content of the past. As wedeconstruct the past we deconstruct the history we write about it.Hence, you or I could deconstruct this book, or White’s Metahistory.Indeed, at the conclusion to Metahistory White himself suggests such

154 Hayden White and deconstructionist history

a course – as he says, the formalism of his model may reflect the pres-ent/emergent ironic stage of human history in which the book waswritten. So, as we deconstruct the past we inevitably deconstruct thediscipline.

White concludes that historians are at liberty to exercise theirimaginations and view Foucault’s four epistemes as essentially sequen-tial periods in which dominant tropes serve to organise knowledge, andwe may choose to argue that the structure of narrative at any time or inany place provides the cognitive conditions influencing how people nar-rated the meaning of their lives to themselves.11 White’s formal modeldoes not entertain such speculations, being epistemologically narrowerand sticking to the issue of how, and in what ways, historians shape andcontour the past through the linguistic, literary and specifically figura-tive forms available. White’s model does not resolve the question ofwhether there is only one fundamental reading or emplotment of thepast available to be discovered by historians. White is concerned notwith the reality of the past as guaranteed by empiricism, but with theemplotment of the impositionalist historian, generating Barthes’ realitye!ect. As I have suggested, White’s formal model does not stop us fromstudying the content of the past, what the past was about, but it castssuch a study in a radically di!erent light. It opens up a new vision ofhow to treat the past at its most basic cultural level, that is, at the levelof narrative.

EVIDENCE

White maintains that the attempt to rediscover or reconstruct the ori-ginal intentions of the author and, therefore, the meaning of the evi-dence is always more di"cult than the forensic problems associatedwith the empirical/inferential method suggest. Beyond the simplest levelof the individual referential statement (that President Madison was 5feet, 4 inches tall), it is the constitution of historical facts as a totalitythat creates their meaning, rather than the discovery or recovery of theessential/original and intentional meaning as constituted by the originalauthor. I have argued that all traces of the past are mediated becausethey are categorised or ordered (narrativised) in some way both in thepast and now – a hermeneutic or interpretative circle. This means thatas we invent an emplotment to transform individual events or state-ments into historical facts, in its turn the emplotment becomes morethan the sum of its parts, and as both White and Ankersmit argue, itis the prefigured emplotment that initially defines the selection ofevidence as well as its interpretation.

Hayden White and deconstructionist history 155

As White says, when historians attempt to explain the facts of theFrench Revolution or decline of the Roman Empire:

What is at issue . . . is not What are the facts? but rather, How are thefacts to be described in order to sanction one mode of explainingthem rather than another? Some historians will insist that historycannot become a science until it finds the technical terminology. . . .Such is the recommendation of Marxists, Positivists, Cliometricians,and so on. Others will continue to insist that the integrity of histori-ography depends on the use of ordinary language. . . . These lattersuppose that ordinary language is a safeguard against ideologicaldeformations of the facts. What they fail to recognise is that ordin-ary language itself has its own forms of terminological determinism,represented by the figures of speech without which discourse itself isimpossible.12

So, as we use language we are subject to its figurative demands, but wealways inject more of ourselves into our partnership with language.White notes that ‘most historical sequences can be emplotted in a num-ber of di!erent ways, so as to provide di!erent interpretations of thoseevents and to endow them with di!erent meanings’.13 The input of thehistorian, therefore, is his/her ability to develop the figurative or meta-phorical nature of the narrative as a form of explanation – to expandthe nature of his/her historical imagination. So, what is history after allthis? Arguably it is the embedding of the real past within the fictive.Only seeing it in this way can we make sense of the past and history?Histories are figurative structures that ‘make sense’ of the past. Doesthis mean, we can emplot any event(s) however we want? As with allsuch issues there is no right answer here. But, as White has concluded,most likely not. This is not because a set of events has an inherentmeaning but because emplotting is a historicist (presentist) action andcertain kinds of narratives in our present culture do not appear to ‘fit’some sets of events. There are plenty of examples. Battles are usuallyemplotted di!erently by the ‘winners’ and ‘losers’. George Bush’s andTony Blair’s ‘war on terror’ will be emplotted di!erently dependent onyour ideological assumptions. The events of the life of George Wash-ington are most likely to be emplotted predominantly as a romance – atleast in the USA where his life is viewed somewhat more romanticallythan in Europe.

However, can we emplot events such as the Nazi Holocaust just howwe want? The dispute between Hayden White and his detractors overthe truthful representation of the Holocaust is an illustration. Assum-ing the de-rhetoricization of the historian’s language then such horrific

156 Hayden White and deconstructionist history

events can only be represented/depicted in certain ways – they can onlyhave one meaning. In other words the inherent story of the Nazi Holo-caust leaves no room for debate over what such an event means.Although White still insists the historian has a choice, he accepts thatcertain kinds of modernist events in all probability do not morallyallow a variety of choices and the Nazi Holocaust was one such harrow-ing event.14 This debate surfaced in the early- and mid-1990s debatesover the representation of the Nazi Holocaust beginning withSaul Friedlander’s edited collection on the representation of the Holo-caust.15 He declared that if we imagine we only bring the past alive aswe write a narrative about it, then we are in danger of doing a substan-tial disservice to the victims of the Nazi Holocaust. Rhetoric (figurationand emplotment primarily) can mean we ‘forget’ its empirical truth.Debates about representation and ways of representation areinappropriate when considering such events so Friedlander argued.

The issue of form is crucial here. Are there only very limited rangesof ways in which the Nazi Holocaust (as well as certain other traumaticor what are deemed to be highly significant nationalist, class or raceevents) can be represented in order that its (their) given meaning is notobscured? Is the rule inflexible that ‘too much imagination’ is simplyinappropriate to the representation of certain empirical events? Hereagain there is no ‘right’ answer. To say that there are only certain waysto think about the past may seem to indicate a degree of intellectualintolerance or, conversely, that ‘right-mindedness’ requires we acknowl-edge the ‘limits of representation’. Ultimately, it may be that where youchoose to stand on this issue comes down less to a debate about therepresentation of the past, than about present politics. It also bearsupon what we understand to be the nature of truth in history. Whateverwe decide, it seems fairly clear that the question of truth is somewhatmore complex than the routine matter of getting the facts straight.

Most historians accept that even if we could revisit and reproduce thepast as it actually was, we would still be interpreting it in our own timeand place, and most likely for our own ideological purposes. No onetoday, apart from the ever-diminishing band of naïve empiricists, ser-iously supports the view that historians objectively recover the past todiscover the truth. In place of naïve empiricism I would argue that wenow have a situation whereby, in practice, most historians seek theirtruth in the past. No matter how pristine is their technical recovery ofthe past in the contextualising and creation of historical facts, I wouldcontend that it is always going to be imposed upon by an ideologicallyaligned and rhetorically constructionist historian. When we interpretthe evidence we contribute to a presumed centre of ‘truth’ by adding

Hayden White and deconstructionist history 157

our interpretation to the weight of existing interpretations. The mean-ing of historical facts so created, in e!ect, changes as historical inter-pretations are continually revisited and the meaninglessness of the pasthas a fresh order imposed upon it through the disciplining of history.The traditional disciplining or domestication of the past through con-stant historical revisionism empties the past of what White calls thesublime: the inherent uncertainty of inexplicable change. The decon-structive consciousness willingly acknowledges the sublime nature ofthe past – its literal meaninglessness, its lack of centre, and its con-sequent lack of truth – while mainstreamers still insist on asking whatwas the past really like and, through dint of professional archivalresearch, still believe they can get ever closer to its truthful reality. Thefact that the truth di!ers for Marxists, post-colonial liberals, feminists,the revisionist right, post-structuralists, or whomever, should warn usof the impossibility of ever getting at the truth.16 Perhaps we could nowaccept that the old distinctions between history and rhetoric, evaluationand fiction were at best only a vogue of the Enlightenment and its nine-teenth-century positivist legacy, the benefits of which were considerablyover-rated?

The denial of the sublime – the What might have been? – in theheadstrong desire for absolute understanding operates most crudely inthe attempt to extract the real, but hidden, intentionality of the authorfrom the evidence. It is in rejecting this improbable process of discoverythat White is at his most insightful. If the past possesses a fictive narra-tive structure, or if it is quite meaningless, then the reconstructionistinsistence on the reality of a historicised context that will allow therecovery of a stable, ordered, preordained and original meaning toemerge from the evidence is destroyed. For White, the e!ective lack oforiginal meaning – which by definition cannot be recovered by recon-structionist contextualism – is important, especially today, because hebelieves it is crucial that history should not be categorised and closedo! for political and ideological purposes by the left, right or centre.Hence his emphasis that our act of organising the evidence as we nar-rativise it immediately and e!ectively closes o! any access either to thegenuine or, for that matter, to other alternative meanings of an event orthought. When we have decided that we know what it means – then thatis what it means.

If the original textualised evidence cannot be trusted to signify whatthe original author intended, then no amount of forensic analysis canrecover what was left out, or de-invent what was invented when theevidence was first created. The element of the sublime that White sowelcomes is lost if we fail to understand the nature of evidence. This is a

158 Hayden White and deconstructionist history

major threat to the reconstructionist mainstream because of the power-ful desire of the Rankean will to know what actually happened. I wouldtend to agree with White that if the historian cannot know the storypresumed to be embedded in the evidence, it is because there are prob-ably an infinite number of stories told in the past and about the past.What the historian must do is locate the di!erent types of stories thatthe evidence will support, which may then be reasonably brought forthas a coherent emplotment informed by the pre-generic plot structure orculturally provided myths that allow the configuring of the facts as astory of a particular kind. The historian may well be convincing, plaus-ible or telling good rather than bad history when the story renderedappeals to the same stock of myths and ideological and methodologicalpreferences shared by the reader. This gives the account or interpret-ation, as White says, ‘the odour of meaning or significance’ andprovides history with its sense of realism.17 What I think White is high-lighting is how this realism is the result of the aesthetic and ethicalchoices made by the historian, even though it is claimed by thosewith reconstructionist or constructionist inclinations to be the resultof years of painstaking, massively in-depth and clinical archivalresearch and interpretation. It should, of course, be possible to haveboth.

As we know, the reconstructionist paradigm of Quellenkritik isintended to provide for the reader not just a verisimilitudinous render-ing of what happened – the odour of meaning – but the legitimateand accurate evaluation of human actions and events. But, as F.R.Ankersmit argues, the historical narrative ‘resembles a belvedere: afterhaving climbed the staircase of its individual statements, one surveys anarea exceeding by far the area on which the staircase was built.18 InWhite’s judgement it is literally from or at this point of view that eachhistorian prefigures his/her interpretative (re)textualisation of the past,by selecting between the several universal tropological strategies,emplotments, arguments with their ideological implications in order toframe the perspective and invest the evidence with meaning but not thetruth. History thus conceived is more like a painting than a forensicreconstruction – an aesthetic appreciation of a past world rather thanthe recovery of its lost reality from the sources composed of individualstatements about past reality. Ankersmit’s contention that we historiansnever employ all the possible range of referential statements available tous about the past means that we select – like colours from a palette –those to be used according to our emplotment decisions (that is, thosethat we judged to be significant). As he says, ‘The results of historicalresearch are expressed in statements’ that can be shown to be more or

Hayden White and deconstructionist history 159

less true, in that they correspond to other statements with the samereferent saying the same things about it; but while narrative interpret-ations are indeed sets of statements, clearly this does not mean that thehistorian’s imposed interpretative narrative structure of interpretationis truthful (or false). Of course, rhetorical constructionism itself, as alinguistic entity, is not immune to the charge that it is also a representa-tion, and thus no more truthful, or likely to be true, than is empiricism.Ankersmit concludes that while individual statements (evidence) maycorrespond to the past in the sense of possessing referentiality, whenthey are collated as narrative interpretations they can then ‘only applyto the past’ and do not ‘correspond or refer to it’.19

The act of interpreting the evidence is not then dictated primarily bythe atomistic content of the past, as reconstructionists would argue; norless does interpretation emerge as jigsawed evidence fits together toprovide the real picture or design. For White the evidence exists in a pre-jigsawed state, a condition that requires the historian to cut and shape itinto a narrative explanation. Historical facts do not ordain interpret-ations; only plot structures do that. Hence, from White’s standpoint,evidence is not inherently tragic, comic, satiric or romantic, and Marx’sfamous aphorism that history occurs the first time around as tragedyand the second time as farce can only be taken to mean that history iswritten by human beings, and the act of narrative interpretation takeshistory out of an unknowable past reality and into the present, ratherthan transmitting us back to the past. This act of creation – literally andmetaphorically – constitutes a plausible historical text rather thanthe past.

Deconstructionist historians do not automatically doubt the truth ofindividual referential statements, nor do they claim that it is impossibleto demonstrate that certain events did or did not happen, that peoplewere not short or tall, that decisions were or were not made, or thatmillions of Jews in the early part of the twentieth century were mur-dered by the German Nazi régime. But the deconstructive emphasis isupon the procedure for creating historical knowledge when we deal withthe evidence. We are aware that we take simple verifiable statements,which we compose into a narrative so that they become meaningful (notnecessarily the same as truthful). This operation is of the historian’simaginative/fictive power. Having suggested that the past is created ashistory by an imposed tropically inspired structure, White neverthelessentertains the thought that it is insu"cient for the historian merely tolearn the language in which the evidence is written, because the histor-ian should penetrate the modes of thought mediated in the tropes. Thisact of penetration is not one of discovery, but one of active narrative

160 Hayden White and deconstructionist history

constructionism on the part of the historian. This is not an engagementof exploration and discovery, but an act of creation.

Convinced as he is that the past has no inherent emplotment, thefiction of factual representation, as White describes it, resides in theprocess of impositionalism. As he says:

Confronted with the chaos of facts the historian must carve them upfor narrative purposes. In short, historical facts, originally consti-tuted as data by the historian, must be constituted a second time aselements of a verbal structure which is always written for a specific. . . purpose.20

It follows that disputes in history depend rarely on facts, but on whatthey mean, and their meaning is determined by the tropic mode of theirnarrative construction – the product of the narrative turn. So, if historyis a primarily narrative construction, what, for White, can be the role ofthe historian’s social theory?

THEORIES OF HISTORY: CONSTRUCTING THE PAST

As White says:

To raise the question of the nature of narrative is to invite reflectionon the very nature of culture. . . . So natural is the impulse to narrate,so inevitable is the form of narrative for any report on the way thingsreally happened, that narrativity could appear problematical only ina culture in which it was absent . . .21

White’s rhetorical constructionist theory of the work of history is builton this insight. He attempts not only to identify the underlying struc-ture of the historical text but, like Foucault, to understand the linguisticforces acting to create the past itself – the modes of thought. He main-tains that historical interpretation ‘might be regarded as what Foucaulthas called a formalisation of the linguistic mode in which the phenom-enal field was originally prepared’.22 White is confirming here thatknowledge is construed through the prefigurative tropes and the tropingprocess. How does the troping process relate to constructionism?

Plainly, White’s emphasis on history as a form of literature defines itas a kind of constructionism. White’s link of the prefigurative power ofthe tropes to Foucault’s conception of the episteme o!ers the possibil-ity of an original and provocative explanatory linguistic model of his-torical change. This model, which White describes as a tropologicalreduction, but which I prefer to call a rhetorical construction, has so farbeen applied in a relatively limited way by historians. The reason is

Hayden White and deconstructionist history 161

three-fold: first, there is a general suspicion of both Foucault and Whitebecause their work questions history as a distinctively empiricist epis-temology; second, and flowing from this, we have the professionalinvestment in the existence of history as a distinct profession; andfinally, and perhaps most tellingly, there is a deep antipathy to anymodel of historical change predicated upon the existence of dominant(and subordinate) tropic prefigurative bases to knowledge – a recon-structionist suspicion of constructionism welded to the irrational fearthat somehow literature will steal the soul of history. For most in themainstream what is taken as linguistic determinism is productive ofempirical uncertainty, and encourages a dangerous moral relativism, ifnot outright nihilism. Any attempt to recover the sublime via rhetoricalconstructionism is a reckless attack aimed at the very heart of theempiricist enterprise.

What White is proposing, following Foucault, is that historians, inwriting history, in e!ect work out a tropically inspired construction.Contrary to this there is the belief – especially popular among Marxistconstructionist historians – that historical experience is redeemable as itactually was only with the intervention of non-linguistic, experience-based social theorising. The term constructionism as I employ it in thisbook refers to history resulting from the imposition of social theorystrategies, models, or covering laws of explanation on the past. As weknow, constructionism, especially sociological and anthropological var-ieties, questions reconstructionist historical inference at the individualevent level, preferring the general level of explanation covering manyindividual events. The use of such historical models has tended toreinforce the notion that historians are able to stand apart from thepast, maintaining the gap between knower and known.23 White holdsthat no historian can maintain such a separation between themselvesand their data because historical narrative will always compromise thisdistinction.

Philippe Carrard’s submission that even the constructionist NewHistory relies upon narrative through which to report the results of itshypothesis testing has important implications. It argues that the mostcomplex application of social theory still depends largely on story-telling ‘to make sense of the world’. He concludes that no amount ofscientific apparatus can disguise the fact that constructionism’s ana-lytical component ‘is still framed by a plot, and this plot retains essen-tial cognitive functions’.24 Not only this of course, but, as White andFoucault insist, the organisation of data into structures is deceptive atthe most fundamental level. Not only is it not an accurate replication ofhistorical events and lost lives, but our impositions are also likely to be

162 Hayden White and deconstructionist history

instruments for the exercise of power – the obvious examples being self-interest in order to maintain history as a professional discipline, or towrite the sort of history that will marginalise and/or exclude minorities,or constitute colonialist and nationalist versions of reality, or, if thehistorian so wishes, to do the opposite of all these. If narrative’s abilityto disrupt as well as establish such power arrangements is recognised,then this can liberate rather than constrain our construction of the past.At its most basic it should remind us never to confuse written historywith referentiality because history is at best analogous to the (a?) past.

The theory of history-writing-interpretation constructed by White is,therefore, a formal model of historical explanation that insists on his-tory as a figurative construction. Every metaphor the historian uses isitself nothing more nor less than a tropic model deployed to re-presentreality, as when Frederick Jackson Turner likened the advance of theAmerican frontier in the nineteenth century to several tides and waves.This description carried with it what was presumably Turner’s deliber-ate intention to create in the reader’s mind the sense of an eternal, pre-ordained and unstoppable natural process. In this particular instance.Turner’s referential signifier was tides or waves of nation-building.25 AsWhite maintains, all historians act constructively when they composethe past. As he says, if we all

treated the historian’s text as what it manifestly was, namely a rhet-orical composition, one would be able to see not only that historianse!ectively constructed the subject of their discourse in and by writ-ing, but that ultimately, what they actually wrote was less a report ofwhat they had found in their research than of what they hadimagined the object of their original interest to consist of. That iswhy I spoke in Metahistory of ‘the historical imagination’ in nine-teenth century Europe and envisioned a ‘poetics of history’ as analternative to the various ‘theories’ of history then circulating.26

In claiming that historical explanation is a form of rhetorical con-structionism, White rejects some of the more extravagant claims forsocial theory history. He is especially critical of the Marxian varietywhich assumes it has discovered in the evidence the actual truth of thepast – the narrative, the story – in class-constructed economic determin-ism. Neither is White, as I read him, suggesting that his rhetorical con-structionist model is to be viewed as a kind of bourgeois or ironiclinguistic determinism or, for that matter, that historians are merely theinstrument of their narratives. What White is doing is far more com-plex, for he is inviting us instead to explore language as the element inwhich everyone (including historians) exists and through which we all

Hayden White and deconstructionist history 163

make sense of our past, present and future. In addition, he reminds usthat our evaluation, interpretation and representation of the events andoccurrences of the past should be, as he says, judged ‘to be “relative” tothe time, place and cultural conditions of its formulation’.27 This kindof contextualism or historicism rests uneasily on the shoulders of mostsocial theory constructionists. Although many constructionists todayare more open to the variety of methodological approaches to historythan are naïve realists, still very few think about how their constructionof the past is changed as they write their narratives. This lacuna bearsfurther examination. Before the application of social theory or hypoth-eses to the data of the past, White holds that we should address theprefigurative or tropic modes in which covering laws, theories andarguments are composed and o!ered. Following the logic of White’smodel this means that literary form, constructed as emplotment, pre-cedes theory and argument, and that all elements or strategies of his-torical explanation are determined by Foucault’s dominant epistemictropic structures.

All six points of the empiricist reconstructionist–constructionistcharter are confounded by White’s argument that before the historiancan bring to bear on the evidence ‘the conceptual apparatus he will useto represent and explain it’ he must first ‘prefigure the field – that is tosay, constitute it as an object of mental perception’.28 This is White’sstriking comment on the process of history-writing as a narrativeimposition by the historian. His formal model o!ers the most satisfyingtheory of how historical narrative works for the historian. His model ispredicated on the belief that to raise the question of ‘the rhetoric ofhistorical discourse is to raise the problem of the nature of descriptionand analysis’, and in this connection he notes the French culturalanthropologist Claude Lévi-Strauss’ comment on traditional history,that in e!ect it has no method ‘uniquely its own, nor indeed any uniquesubject matter’. The important point that White makes here is thathistorical writing

must be analysed primarily as a kind of prose discourse before itsclaims to objectivity and truthfulness can be tested. This means sub-jecting any historical discourse to a rhetorical analysis, so as to dis-close the poetical understructure of what is meant to pass for amodest prose representation of reality.29

White is saying that written history should be categorised primarilyaccording to the way in which it describes its object of study rather thanthe empirically derived explanatory mechanisms of analysis it appliesto the evidence (collection, colligation, comparison and verification).

164 Hayden White and deconstructionist history

Historical writing explains because of the way it is put together – thecontent of the form. To understand fully the role of narrative in thewriting of the past, and its aesthetic, cognitive and moral content, it isnecessary briefly to outline White’s formal model of the historicalimagination which in turn may give us a significant insight into thecharacter of historical explanation.30

HISTORY AS NARRATIVE

White’s model is best viewed as a grid of relationships, or elective a"n-ities as he calls them, between the levels or strategies of historicalexplanation that every historian utilises in reconstituting the past:

Following the four main kinds of trope, which act as the foundation forall historical interpretation, are four kinds of explanation in three tiers,namely four emplotments, associated with four types of argument andfour ideological positions. The historian may emplot his/her narrativein one of the four modalities available – the romantic, tragic, comicor satiric. These then influence his/her appeal to one of four modesof argument – formist, mechanistic, organicist or contextualist; andfinally, his/her selection of plot and argument has ideological implica-tions that negotiate the union of the prior aesthetic and cognitivestrategies of explanation. The ideological implications are anarchism,radicalism, conservatism or liberalism.

As an aid to understanding how history is written, this model is bestsummarised by White in Metahistory:

These a"nities are not to be taken as necessary combinations of themodes of a given historian. On the contrary, the dialectical tensionwhich characterises the work of every master historian usually arisesfrom an e!ort to wed a mode of emplotment with a mode of argu-ment or of ideological implication which is inconsonant with it. Forexample . . . Michelet tried to combine a Romantic emplotment anda Formist argument with an ideology that is explicitly Liberal. So,

Trope Emplotment Argument Ideologicalimplication

Metaphor Romantic Formist AnarchismMetonymy Tragic Mechanistic RadicalismSynecdoche Comic Organicist ConservatismIrony Satiric Contextualist Liberalism31

Hayden White and deconstructionist history 165

too, Burckhardt used a Satirical emplotment and a Contextualistargument in the service of an ideological position that is explicitlyConservative and ultimately Reactionary. Hegel emplotted on twolevels – Tragic on the microcosmic, Comic on the macrocosmic –both of which are justified by appeal to a mode of argument thatis Organicist, with the result that one can derive either Radical orConservative ideological implications from a reading of his work.32

Clearly then, although the model must inevitably be laid out formally, itis not rigid or absolute in the relationships it o!ers. But, like Foucault,White is claiming that all three surface tiers of narrative are under-pinned by their tropic or figurative structure, which is (to remind our-selves) the process we engage in as we describe to ourselves and othersthe relationships presumed to exist between objects, texts, events andcontexts. This is the metahistorical level of the model to which Whiterefers as that set of presuppositions which ‘is nothing but the web ofcommitments which the historian makes in the course of his interpret-ation on the aesthetic, cognitive, and ethical levels’.33 This basic level ofconsciousness is ‘that of language itself, which, in areas of study such ashistory, can be said to operate tropologically’ in order to prefigure ‘afield of perception in a particular modality of relationships’.34 In otherwords, this troping process creates an imagined context through which,for illustration, metonymic (part-to-whole) or synecdochic (whole-to-part) relationships are established between events and things. Becausehistory is literature it can apprehend its data only through the dictatesof its narrative form; the figure(s) of speech we impose on the data actto sanction the very nature of our historical understanding. The tropes,as representational models, thus preshape our descriptions of the data,preceding and prefiguring the emplotment, argument and ideologicallevels of our historical narratives.

Troping means using metaphors to imply meaning and explain eventsby altering our perceptions, forcing us to look again at objects andconcepts from the perspective of something di!erent – signification andresignification. Each of the four tropes is defined according to its par-ticular rhetorical and, therefore, explanatory function. We relate eventsand human actions not according to some wholly extrinsic situation,but through language, and specifically how language operates inrelating parts to wholes and vice versa. As White says:

Irony, Metonymy, and Synecdoche are kinds of Metaphor, but theydi!er from one another in the kinds of reductions or intergrations[sic] they e!ect on the literal level of their meanings and by the kindsof illuminations they aim at on the figurative level. Metaphor is

166 Hayden White and deconstructionist history

essentially representational, Metonymy is reductionist, Synecdoche isintegrative, and Irony is negational.35

Following Vico, White takes this troping process to constitute theinescapable prefigurative act of historians and all those who write nar-ratives. White, like Foucault, is suggesting that the mechanism of writ-ing history operates at the subterranean level of language and humanconsciousness – the prefigurative act being divined in and through ‘thedominant tropological mode in which it is cast’.36 Even if the aimremains the stated attempt to recapture what actually happened in thepast, even then we must first ‘prefigure as a possible object of knowl-edge the whole series of events reported in the documents’.37 This kindof contextualisation is not just a narrative version of empiricism (whichendorses the link between statement and referent), it is rather a meansof understanding that precedes empiricism as well as all other phil-osophies of history. The result is that as historians we do not objectivelyreport the past, we create it as we deploy language to define the con-cepts used – not merely to identify our objects of study, but also tocharacterise the kind of relationships (troping) we see (imagine)between them. History is validated not solely by the appeal to pastreality, but also by how that reality is written. As one commentatorwrote, history after White cannot be judged as if it were itself outsidehistory – some natural object of reality.38

For White, the important point is that in metaphor, metonymy,synecdoche and irony

language provides us with models of the direction that thought itselfmight take in its e!ort to provide meaning to areas of experience notalready regarded as being cognitively secured by either commonsense, tradition, or science. And we can see that in a field of studysuch as history ‘interpretation’ might be regarded as what Foucaulthas called a ‘formalisation’ of the linguistic mode in which thephenomenal field was originally prepared . . .39

Metaphor thus identifies thought as representational in which the simi-larities between objects are stressed, metonymy signifies by reducing anobject to a part or parts; synecdoche works the other way by integratingobjects, emphasising their similarities or essences, and to tropeironically is to negate literal meaning.

If we use the example of Frederick Jackson Turner’s history of theAmerican frontier to illustrate the troping process, one of his claims isthat the ‘free land’ of the frontier possessed the power to assimilate orAmericanise pioneering settlers. Read as a metaphor, ‘free land’ is

Hayden White and deconstructionist history 167

defined as ‘the line of most rapid Americanisation’. If read as a meto-nym, Americanisation is reduced to its most significant part, namely theexistence of ‘free land’. If read as a synecdoche, ‘free land’ signifiesthe essence of Americanisation. If read ironically, ‘free land’, stated asthe literal truth of the Americanisation process, would be negated by thecontext created by the historian that there was none and Americanisa-tion thus never occurred. Such rhetorical constructions, as White warnsus, can act as particularly e!ective political encodations serving theexercise of power – history as ideology. To this end, when written ashistory, Turner’s frontier thesis is ultimately integrated into hisdefinition of America by emphasising its essential nationalisingcharacteristic, which he defines as ‘the existence of an area of free land’.

These tropologically determined or prefigured strategies of explan-ation allow us as historians an inventory from which to choose in orderto construct narrative explanations or interpretations of the past – evenfor those among us whose desire it is to know what actually happened,or who wish to confirm our own particular constructionist explanation.This logic is translated by White into his argument that the actualchoice of emplotment is ultimately the product of the episteme in whichit is written, because it draws upon contemporary cultural referencepoints, or as he calls them archetypal story forms, which themselvesaccord with the dominant (and/or subordinate) myths and needs ofsociety as cast at the tropic or metahistorical level from which ourwritten history emerges. The American frontier experience, as emplot-ted by Frederick Jackson Turner in the 1890s, chimed with the needs ofthe dominant entrepreneurial social formations at that time to create auseful heroic pioneer identity for themselves. Hence, in what to us in ourepisteme may appear to be a rather implausible notion, that the reces-sion of the frontier provided the forces by which the composite nation-ality of the United States was created, his emplotment neverthelessserved at the time important nationalistic ends in a period of culturalcrisis. It did this by providing the story of the pioneer-hero – the arche-typal individualistic American transmuted into the entrepreneur –whose frontier-inspired powers would eventually resolve the manyproblems that beset America in the turbulent 1890s. For Turner, historyitself became the ultimate metaphor.

The context of Turner’s history supports White’s contention that themost convincing stories told by historians will be those that echo con-temporary cultural myths and beliefs, as they conform to the recognis-able and dominant Western story-types of romance, tragedy, comedyand satire. Hence, the story fashioned by the historian tropically pre-figures the way in which the traces of the past are connected together

168 Hayden White and deconstructionist history

through the kinds of figurative reductions or integrations I noted above.The emplotment, for example, is selected as a result of the historian’sconception of the power of action of the hero or protagonist over his orits environment. Because, in White’s view, no set of historical circum-stances is inherently tragic or comic, the emplotment is written by theimpositionalist historian as a consequence of imposing a judgement onthe nature of the sequence of events encountered by the designatedhero or protagonist. The emplotment becomes the vehicle of his/herhistorical explanation.

In describing events as a romance, the historical emplotment is iden-tified by the historian imagining the power of the historical agent/heroor protagonist as ultimately superior to his or its environment. Historyemplotted as a romance unfolds as a quest with final success, redemp-tion or transcendence assured. In Western culture this history is usuallydescribed as a journey, a struggle, with eventual victory over adversityfor the hero or protagonist, whether nation, state, individual, class, orwhatever. Emplotment as a satire is the polar opposite to romance inthat the agent/hero or protagonist is imagined by the historian as beinginferior, a captive of their world, and destined to a life of obstacles andnegation. Tragic emplotments are similar to romance in as much as theyidentify the hero or protagonist as eventually thwarted by fate or theirtragic personality flaw. The end result is failure, defeat or death. Thefate of the bourgeoisie in Marxian history, for example, is usually castas a tragic emplotment. In a comic emplotment, movement is imaginedfrom obstruction to reconstruction, and the historian always hopes forat least a temporary victory over circumstance for the hero or protagon-ist through the process of reconciliation. Festivities at the closure ofsuch historical narratives usually celebrate the coherence and consensusachieved with other men, women, races, nations or classes by the heroicfigure.40

In addition to the narrative level of emplotment of events, there isanother level on which the historian tries to explain ‘the point of it all’or ‘what it all adds up to’, which is explanation by formal argument(formist, mechanistic, organicist and contextualist arguments)41

Explanation by argument means that as historians we o!er to ourreaders more or less convincing but always commonly accepted laws ofhistorical change or human behaviour upon which we all draw toexplain events. The arguments we use relate events, people and actionsby the appeal we make to our own thinking processes of dispersion orintegration. A formist argument, for example, identifies the unique,atomistic or dispersive character of events, people and actions in thepast. Such an argument permits us graphically to represent vivid

Hayden White and deconstructionist history 169

individual events from which it is possible to make significant general-isations. Thus the winning or losing of great battles or civil wars consti-tutes the origins of great historical change, or the special lives of greatmen and women are taken to signify the nature of historical change; aclassic form of the latter is the genre of born in poverty to becomepresident, or overcome social disadvantage to rise to statesman, or con-quer prejudice to emerge as a race leader. Organicist arguments, whichare integrative, allow us to identify past events, people and actions ascomponents of a synthetic process in a microcosmic–macrocosmic rela-tionship whereby a single event or an individual is just one elementamong many – a factor that goes into making for complex historicalchange. Mechanistic arguments tend to be reductive rather than syn-thetic, usually cast in the form of an equivalent part–part relationship,and with which we identify events, people and actions as subject todeterministic extra-historical laws. As the term suggests, contextualistarguments are moderately integrative and are employed by those histor-ians wishing to identify events, people and actions in the past by theirpresumed connections to others in webs of colligatory relationshipswithin an era, or within a complex process of interconnected change.

A little thought produces many examples of rhetorical construction-ism operating as explanatory argument. White himself o!ers theexample of the relationship between Marx’s base and superstructure asa classic mechanistic ‘law’. According to this ‘law’, transformationsin the economic base ultimately determine changes in the social/ideological superstructure, but the reverse does not occur (part to part).Thus, the tragedy of Marx’s mechanistic explanation of historicalchange lies in the failure of the heroic proletariat to successfully over-throw their bourgeois oppressors in Europe and America in the latenineteenth century. Another rhetorical construction could have it dif-ferently, as in the case of the Russian Revolution which might beimagined as a successful romance of proletarian victory. So, historicaldisputes are less about what did or did not actually happen, andmore about how we emplot or invoke ‘laws’/social theories to explainthe past.

There is, however, the third level in White’s formal model of histori-cal explanation, and this final strategy of explanation is the ideological.As White’s description suggests, the ideological is the moral implicationof our choice of emplotment and argument. Accordingly, in the histori-cal narrative the ideological level projects ‘the ethical element in thehistorian’s assumption of a particular position on the question of thenature of historical knowledge and the implications that can be drawnfrom the study of past events for the understanding of present ones’.42

170 Hayden White and deconstructionist history

In this way White acknowledges the presentism of history and its con-struction as a contemporary cultural discourse. He acknowledges thatno historian can stand aside from history and suspend his/her capacityfor or exercise of moral judgement. The question of ethics and historyhas been an important development from the 1990s up to the present.What is the role of the ethical turn in history? It seems hardly tenable toargue that historians do not hold ethical positions, and it is only slightlyless convincing to say they can suspend them when they write history.Indeed, ethical choice is an important strategy for explanation in his-tory. It seems reasonable to say that history is as ethical as any histori-cist, constructed, and cultural narrative can or can’t be. In other wordsethics and the representation of the past exist in the same universeof narrative making. If historians o!er a value added to the world ofethical living, I assume it is a result of their sincerity and prudence inthinking about history as an aesthetic form of representation ratherthan simply just doing it – in other words, rethinking the empirical-analytical epistemological model with its commitment to meaningthrough what happened and knowable agent intentionality, with a morecomplex and mediated concept of history as a second order philosophy.If aesthetics precedes history, so ethics precedes aesthetics.

If truth defined as ‘real meaning’ is untenable in a post-structuralistworld – not in the denial of past reality but in terms of understandingthe meaning of ‘the other’, the di!erent – then the moral worth of ‘the-past-as-history’ can only be located in the narratives we construct aboutit ‘as if ’ they mean something we ultimately find desirable. If we can’tknow the true meaning of the past despite knowing what happened,then the historian who wishes to have an answer to what it all mightmean can only begin with his or her own attitude toward ‘the other’. AsHayden White, Emmanuel Levinas and Frank R. Ankersmit remindsus, ethics precede truth, and knowing ‘things’ is no guide to an ethicallife.43 In an epistemologically sceptical universe it might be argued wedon’t look to the past, much less history, as a source for engaging withand understanding ‘the other’. We create them both as we moralise.

The nature of our impositionalism means that there are no disinter-ested historians. White makes this very clear when he defines ideologyas a set of ‘prescriptions for taking a position in the present world ofsocial praxis and acting upon it (either to change the world or tomaintain it in its current state)’. He postulates four basic ideologicalpositions (borrowing from the German philosopher Karl Mannheim) –anarchism, radicalism, conservatism and liberalism. These four basicpositions all claim to be rational or scientific in the modernist sense ofhaving to make such claims to the master narrative of science in order

Hayden White and deconstructionist history 171

to be heard.44 The ultimately ideological nature of empiricist history liesin the way in which it attempts to make us all read its works as if theywere realistic – this is the truth of the matter, or we really must face thefacts – and thus we can respond only in certain ways.

In spite of the seemingly pre-eminent or determining power of thetroping process, White is unsure that ideology is absolutely the result ofform, that is, that the trope ultimately determines ideological posi-tions.45 At one point he says that the four primary ideological positionsare less influenced by troping (between reduction, dispersion and inte-gration) than by the historian’s attitude towards what is desirable in thenature and pace of social change, that is, that the historian has a freemoral choice unencumbered by the power of figuration, and that,therefore, all four ideological positions in their di!erences as to thedesirability of its direction and pace may be autonomous. Briefly, con-servatives are most suspicious of change, the others less so. Conserva-tives oppose rapid change by supporting the evolutionary elaborationof existing social institutions. Anarchists demand rapid, perhaps evencataclysmic, social change in order to establish a new society. Liberalsprefer the fine tuning of society to secure moderately paced socialchange, while radicals welcome imminent social change, but unlikeanarchists are more aware of what White calls the ‘inertial pull ofinherited institutions’ and are, therefore, much more exercised by themeans to e!ect change than are anarchists. All four positions thus carrypreferences for the distribution of power and the criteria by whichpower is exercised. What this means is that the ideology of the historianis refracted through the history he/she writes, and as to the question ofultimate determination – trope or ideology – the answer is always likelyto remain disputed.

Nevertheless, White’s model is clear that the explanatory historicalnarrative depends not on the facticity of events for its functioning, butrather upon creating a story, deploying arguments, and taking up moralpositions that the reader can follow and understand in shared con-temporary cultural terms. Precision in the generation of facts is literallymeaningless unless and until those facts (as propositions or eventsunder a description) are turned into stories, explained further witharguments, and o!ered as sustained and coherent ideological positions.Louis Mink, quoting the literary critic Barbara Hardy, makes animportant point in this respect:

Narrative, like lyric or dance, is not to be regarded as an aestheticinvention used by artists to control, manipulate and orderexperience, but as a primary act of mind transferred to art from

172 Hayden White and deconstructionist history

life. . . . More important than the artifices of fiction are the qualitieswhich narrative shares with the story-telling of lived experience: ‘forwe dream in narrative, daydream in narrative, remember, anticipate,hope, despair, believe, doubt, plan, revise, criticise, construct, gossip,learn, hate, and love by narrative’.46

Now, although the level of understanding to which narratives aim is forMink a primary act of mind, he draws back from the final logic ofHardy’s position, and here White agrees with his conclusion that

Stories are not lived but told. Life has no beginnings, middles orends: there are meetings, but the start of an a!air belongs to thestory we tell ourselves later, and there are partings, but final partingsonly in the story. There are hopes, plans, battles, and ideas, but onlyin retrospective stories are hopes unfulfilled, plans miscarried, battlesdecisive, and ideas seminal. Only in the story is it America whichColumbus discovers, and only in the story is the kingdom lost forwant of a nail. . . . So it seems truer to say that narrative qualities aretransferred from art to life. We could learn to tell stories of our livesfrom nursery rhymes, or from culture-myths if we had any, but it isfrom history and fiction that we learn how to tell and to understandcomplex stories, and how it is stories answer questions.47

In spite of this anti-narrativist assertion by White, we must at least bepermitted to ask once more, is history really only the fiction that histor-ians tell as they order the evidence, or is there not a cultural resonancebetween history as it is lived and as it is told?

CONCLUSION

According to the formalism of White’s tropic model, history is a pro-cess of continuous intertextual reinscription composed and conductedby the historian – it is primarily an act of literary creation. Because thecharacter of historical interpretation resides in its narrative structure,historical knowledge is generated by the constant debates between nar-ratives (interpretations) rather than the primeval, uninscripted anduncontextualised traces of the past. F.R. Ankersmit makes the pointwell with his claim that all we historians have is the ‘intertextual’ inter-play between our historical narratives.48 All debates in history – whostarted the Cold War, how successful were the Chartists in achievingtheir aims, to what extent was the recession of the American frontierculturally significant in American history? – are debates between com-peting narrative interpretations. Moreover, because the historical

Hayden White and deconstructionist history 173

imagination itself exists intertextually within our own social and polit-ical environment, the past is never discovered in a world set aside fromeveryday life. History is designed and composed in the here and now.

Whether the impositionalism of the historian is ultimately structuredby ideology or trope is impossible to determine by either the evidence orcriticism. The historian’s position may be morally and/or rhetoricallyinflected. What is important, however, is that the historical imaginationis both a lightning rod and a conductor of culture, past and present.Alongside the tropically prefigured epistemic model of cultural forma-tion provided by Michel Foucault, White’s formalist theory of histori-cal narrative o!ers a morphology for the study of the past. Agreeingwith White that historical understanding as derived from evidence liesnot at the level of individual referential statement but in its emplottedarrangement, leaves open the question that in the past there may havebeen dominant and tropically constituted epistemic narrative structuresthat mediated the nature of historical change. This leads me to theconclusion that the real poverty of empiricism resides in its strenuousrefusal to acknowledge the power of figuration in the narrativisation ofthe past as exercised then as well as later by the historian. The implica-tions of this contention for the deconstructive consciousness forms thebasis of my summary in the concluding chapter.

174 Hayden White and deconstructionist history

9 Conclusion

INTRODUCTION

In this book I have asked how can what the content of the past meansbe influenced by the form of its presentation? I pursued this by askingfour key questions about epistemology, evidence, social theory and nar-rative. My thoughts on these questions were facilitated by describingthe three dominant current approaches towards the historical enter-prise, the twin tendencies of empiricist reconstructionism and socialtheory constructionism, and what I have characterised as decon-structionism. Each o!ers a distinctive methodological orientationtowards the four questions. As we have seen, the three methodologicalorientations not only signpost the complexities and varieties of his-torical method available today, but they also reveal the fundamentaldi!erences among historians over the nature and roles of objectivity,explanation, truth, description and meaning in historical understanding.

The deconstructive historical consciousness does not suggest that,because we cannot have a direct and unalloyed access to the actuality ofthe past, history must be characterised as either a purely mental orpurely linguistic entity. We can still talk about the past and what wethink happened in it. But what I have proposed is that, despite theclosest scrutiny of the evidence, and in the absence of a direct cor-respondence to the reality of the past, the way in which history is inter-preted and reported as a narrative is of primary importance to theacquisition and character of our historical knowledge. While it isunproblematic to accept that the reality of the past once existed, it isalso reasonable to argue that we cannot gain access to it solely or evenprimarily through the empirical method. Deconstructionist historiansdoubt whether we can really know the past as it actually happenedby following the six points of the mainstream charter. This is notanti-history, but is a conception of history as what it palpably is: a

self-conscious narrative composition written in the here and now thatrecognises its literary form as its essential cognitive medium, and notmerely its mode of report. Following on from this I have asked if thepast itself was constituted as a narrative by people in the course of theirlived experience, a process that may be imprinted in the evidence. I have,therefore, questioned the degree to which historians discover the past,or can choose to write either the past, or a past, in the shape of thestory, or a story. This led to my contention that we must always dis-tinguish the past from history. My understanding of the importance ofthe form of history to the nature of historical change, as well as thehistorian’s account of it, I examined by reference to the strategic com-bination of Michel Foucault’s conception of the epistemic/tropic socialinfrastructure, and Hayden White’s formalistic model of the historicalimagination.1 I will now move to the implications for the discipline ofhistory of this rethinking of its nature.

EPISTEMOLOGY

In our postmodern millennial climate of doubt and dispute manythings once held to be certain are questioned and, among many otherthings, the empirical grand narrative is not exempt from this interroga-tion. Specifically, deconstructionist history focuses the growing objec-tion of many historians and critical thinkers on Western thought’smodernist belief in the representational or correspondence theory link-ing the word and the world.2 Although the philosopher of history LouisMink has argued that to explain historical events does not now dependon any supreme model of explanation, whether empiricist, covering lawor, for that matter, deconstructionist, the main development in the phil-osophy of history in the past generation has surely been the proposalthat history’s primary cognitive device may reside in its power of narra-tion.3 Until relatively recently few history texts were so epistemologic-ally self-referential as to draw deliberate attention to their rhetoricalform, preferring instead to foreground the presumed reality behind it.Consequently, texts like Emmanuel Le Roy Ladurie’s Montaillou, CarloGinzburg’s The Cheese and Worms, Natalie Zemon Davis’ The Returnof Martin Guerre, Simon Schama’s Dead Certainties and his Landscapeand Memory are regarded as exemplars of a new historical genrebecause they draw attention to themselves either through their contentas studies of the trivial, the anecdotal, the apparently historicallymarginal, or in respect of their form, as illustrations of where historytransgresses the border into fiction through the particular ways in whichthey organise the content of the past.4

176 Conclusion

Books like these do not purport to refer to the truth of the past whichwe can know objectively through the forensic study of the evidence. AsNatalie Zemon Davis said, her book was intended to question the pointat which history ceases to be reconstruction and becomes invention,concluding that she chose to advance her arguments ‘as much by theordering of narrative, choice of detail, literary voice, and metaphor asby topical analysis’.5 It is at this point that the historian disobeys thetraditional rules by replacing the authority of the source with the formof its organisation. Through this the past is liberated because it is nolonger the captive of the reconstructionist–constructionist intellectualestablishment. As has been argued, postmodern or deconstructionisthistory converges no longer on the past as such, but on the disjuncturebetween pastness and presentness. The historian from the conservativereconstructionist mainstream is an empiricist who believes in a know-able historical reality independent of the mind of the historian – subjectand object are separated just as mind and knowledge are presumed tobe. The text is there merely to convey, as explicitly as possible, themeaning of the past. However, all texts conceal the past through theideologically inflected intentionality of the author-historian. Instead ofbeing there but not examined, the text is now the focus for our study ofthe past. This does not mean, as Derrida would have us believe, thatthere is nothing beyond the text, for the text is not the end of history, itis the beginning. Frank R. Ankersmit has made the most consistentcase throughout the 1990s and into the first decade of the present cen-tury for the ‘aesthetic turn’ in the study of history. He has argued thathistorians ought to be conscious of the aesthetic nature of history, notto trade rational thought or empiricism away, but to extend the promiseand potential in their understanding of the past as history. For Ank-ersmit, every historian has to engage with one fundamental epistemo-logical issue. This is whether the meaning of the past is resolutelydetermined through its empirical content or the form of its representa-tion, or some unfathomable blend of them. If you believe historyshould be de-rhetoricized, then its aesthetics are not epistemologicallynoteworthy. Presumably, you will think the fact that history has to bewritten is in no way an obstacle, as the practical realist philosopher ofhistory C. Behan McCullagh says, to o!ering ‘credible, intelligible andfair history’.6 If, however, you disagree with McCullagh, then you maybelieve that exploring history’s aesthetic nature becomes urgent andessential in order to grasp what credible, intelligible and fair history isreally like.

The deconstructionist historian is always uncertain about history.Although it may be di"cult to overcome old historians’ habits, it is not

Conclusion 177

essential to establish either teleological or total explanations. In objec-tion to the certaintist and naturalistic myths of reconstructionist-contextualist empiricism, or the empirically founded social theorisingof constructionists, the deconstructive awareness of history’s inventednature does not allow that the past and history are the same. That thereare no natural links, only assumed epistemological ones, between realevents in the past and the way in which we describe them is the force ofF.R. Ankersmit’s contention that history is best viewed through theperspective of a narrativist rather than an epistemological philosophyof history.7

The past does not dwell ‘out there’, with an existence independent ofthe historian and his/her use of language. If this is so, say reconstruc-tionists, how can we tell good from bad history and without the tran-scendent benchmark of empiricism, the anchors of factualism and theforensic study of the evidence, are we not adrift in a sea of relativism?How can we trust the history we read? Posing such questions revealsmuch about the methodological limits to empiricism. Such worriesannounce the reconstructionist’s uncited investment in the trope of thereal. Many, hopefully most, historians today would not accept Ger-trude Himmelfarb’s argument, derived from philosophical realism, thatour impositionalism must mean that we construct the past without anysense of what is morally right just because we do not know what is true.This is a bleak argument that does not do justice to the dissenting andquestioning nature of much historiography.

The challenge mounted by Barthes and Derrida to the referentialrelationship of the word and the world has formed a part of the broaderobjections of the likes of White, Megill, LaCapra, Jenkins, Ankersmit,Kellner, Rüsen and Foucault to the traditional paradigms. UnlikeHimmelfarb, these critics refuse the Baconian belief that we can gainentrance to the Real World of the Past through the bits of reality scat-tered through the archives. Instead of the presumption of getting closerto the evidence, and hence through it to the truth of what actuallyhappened, an alternative historical understanding is o!ered. Thisdeconstructive epistemology recognises the existence of the reality-e!ect rather than the fantasy notion of historical truth, denies that wecan discover the intentionality of the author, accepts chains of inter-pretative signification rather than recoverable original meaning, refusesthe seductions of the easy referent, disputes the objectivity of thehistorian as he/she works within the figurative structure of narrative,accepts the sublime nature of the past imagined as a sense of ‘theother’, and admits that the form and content relationship is more com-plex than many in the twin main tendencies often allow.

178 Conclusion

In 1990 the Africanist historian Elizabeth Tonkin, in pursuit of themultiple voices of the past, claimed no longer to use the word ‘history’,preferring the term ‘representations of pastness’ because, as she under-stands the nature of historical knowledge, histories are simply chains ofwords ‘either spoken or written, ordered in patterns of discourse thatrepresent events’. She continued: ‘Arguments and opinions too areforms of words. When we grasp a historical fact or interpretation, wehave ourselves made an extremely complex set of interpretations to doso.’8 The deconstructive consciousness makes us epistemologicallyaware that the way in which we metaphorically prefigure, organise,emplot, explain and make moral judgements about the past is our onlyaccess to it. If the reader is persuaded by the general approach ofthe historian, accepts his/her practices and his/her methodologicalorientation, then no detectable gap exists between reader, text andunderstanding. If the linguistic customs and conventions employed bythe historian, and shared with the reader, are those of reconstruction-ism, then again there is no detectable fissure between the historicalaccount and what is believed to have actually happened. Either way, thehistorian has to tap into the epistemological assumptions of his/herreadership. If historical explanation does not work within one or moreof the common intellectual currencies of the present in which it iswritten, it can never explain anything to anyone.

White’s primary contention that history is a rhetorical constructionof the historian, as much invented as found, means the past as it actu-ally happened must ultimately be unknowable. Rather than being ableto grasp the past’s real meaning as it is objectively filtered through themesh of the evidence by the historian, and which is then absorbed bythe reader, deconstructionist history stresses the interactive and imposi-tionalist role of the historian, so that whatever knowledge we have ofthe past is provided not only by the past itself, but by the tropically pre-figured, emplotted, argued and ideologically positioned narrative of thehistorian. The question remains, however, in spite of White’s insistencethat the past is not inherently emplotted, that if Foucault’s trope–episteme link does exist, then is there an ultimate determination inhistory to be located as historians tell stories about the past? The uni-verse of contemporary cultural and linguistic experience shared by his-torians is the ultimate epistemological constraint on not only what canbe written as history, but how it is written.

Conclusion 179

EVIDENCE

I do not doubt that the past once existed, and that evidence of itremains in our present. However, the intractable epistemological prob-lem of never knowing the past as it actually was, because all we can dois infer meaning through its traces, emphasises the need to readdressthe mainstream working assumption of an adequate correspondencebetween evidence and the truthful knowledge of the past. This doubtinforms my answer to the question: what is the character of historicalevidence and what function does it perform? The naïve empiricist pos-ition, which necessarily assumes the existence of objective and judicialhistorians separated from the evidence, who keep their minds free froma priori assumptions, who avoid questions that beg answers, and whoapply to the evidence forensic procedures for the critical evaluation ofthe evidence, is now more widely disputed than ever before.

I have tried to show how it is the reconstrutionist belief that ‘truth’corresponds to past reality via the mechanisms of referentiality andinductive inference (which I summarised as the six points or principlesof the empiricist charter), which prompts historians like McCullagh,Elton, Standford and Marwick to argue that historical truth can bediscovered by recovering the intentionality of the author of the evi-dence. This position, as developed but not rejected by reconstructionistmoderates like Kloppenberg, Appleby, Hunt, Jacob and Gordon,means that they seek out the story that most faithfully represents thetruth which can and must eventually be found in the past through theundisputed factual detail of past events. While they acknowledge narra-tive as the medium for this empiricist reconstruction of the past, itspower to invent is strenuously denied. Historians serve the evidenceunder all circumstances. The same thinking applies to deductive socialscience theorising, so much so that both deconstructionism and con-structionism are viewed as unnatural acts going against the grain ofevidence-based historical methodology.

We should by now be familiar with the Elton argument that history iswhat results from the evidence as interrogated by the disinterested andindependent historian who, in asking appropriately framed questions,remains particularly sceptical about paradigms of explanation pro-posed by both social theorists as well as deconstructionists. Among theconsequences assumed to flow from this method is not only the gener-ation of incontestable facts derived from the discovery of the intention-ality of the author of the evidence, but also the emergence of cleardistinctions between history and value, fact and fiction. Constructionisttheories of history are thus rejected by hard-core reconstructionists

180 Conclusion

because of the injuries they do to the relics of the past as they are forcedinto strange shapes dictated by the needs of the hypotheses to be testedand, as Elton has it, invariably confirmed. For both conservative andmainstream historians the ultimate defence against what is proclaimedas the relativism of deconstructionism resides in the practice ofQuellenkritik – the technical and detailed study of the sources throughthe process of verification, comparison and colligation.

In the past half century, however, under E.H. Carr’s influential visionof history that the past is always imperfectly encountered, most prac-tical realist or moderate mainstream historians have had little di"cultyin accepting what they call the provisional nature of their interpret-ations and that proof and truth do not exist in history. This translatesreadily into historical revisionism. The provisional nature of all his-torical interpretation rests on the continuous process of discoveringnew evidence, treating it with increasingly sophisticated mechanisms ofanalysis and conceptualisation, and constantly recontextualising it sothat, for example, the evidence of Empire becomes, for the next gener-ation of historians, the evidence for a new post-colonial interpretation.Although this happens, the available evidence is, of course, still believedto provide a window on past reality. While new evidence always o!ersnew windows, revisionist views will still correspond to the reality foundbeyond the new windows. Revisionist constructions placed on the his-torical facts do not, therefore, destroy the knowability of past reality.Even given Collingwood–Carr relativism, the ultimate claim of recon-structionism remains intact, that the past is knowable through the evi-dence, and remains knowable even as it is constituted into the narrative,for it is then that we get to the story or true description of the past. Formainstreamers the aim remains to get ever closer to the truest possibledescription.

My response to this vision of the existence of a knowable historicalreality has been to question the empricist belief that the explanatoryvalue of the evidence increases the more we turn up the magnificationon the Quellenkritik microscope. I have submitted that it does not fol-low that the closer to the evidence we get, the more we see of the truth. Ihave not disputed that the correspondence of evidence with realityworks reasonably satisfactorily at the basic level of the single sentencesupported by the evidence (US President Abraham Lincoln was shot on14 April and died in the early morning of 15 April 1865). But, suchcorrespondence does not exist when we shift a gear to the level ofinterpretation through the imposition of an emplotment or argument(Abraham Lincoln was assassinated before he could put his reconstruc-tion plans into e!ect). It bears repetition: the historical narrative is not

Conclusion 181

the past, it is a history. While it may be possible to demonstrate a strong,even a probable correspondence between a single statement about thepast and a single piece of evidence, su"cient to generate a factualstatement, to then translate this inductive ‘truth’ to a whole historicalinterpretative narrative, so as to recover the past as it actually was, is aflawed practice.

That evidence does not correspond with past reality at the interpret-ative level makes discovering authorial intentionality particularlyproblematic. To argue, like Elton, that the detached historian canunderstand the intentions of people in the past by asking why did theevidence come into existence, displays an extraordinary level of artless-ness. The innate imprecision of inductive inference ought to warn o! allhistorians from such a belief, but clearly does not. Claiming to seek thetruth inferentially may be psychologically and professionally satisfying,but it is always intellectually dangerous, and no more so than whensomeone believes they are getting ever closer to it, or have found it. Thesteel trap of certainty then closes over the inquiring mind. Always todoubt and to question is to welcome the inherently discontinuousnature of the writing of the past – a position that may lead to a moreencompassing form of historical analysis that is less likely to close o!the marginalised and the ‘other’.

The most thoughtful and mature appreciation of what constitutesmainstream history today is to be found in the work of the historianslike Appleby, Hunt and Jacob who argue that their commitment to thestudy of the evidence is to discover the truth of (in their case) a plural-istic and multicultural American history that will necessarily reinforce(what they choose to believe is) America’s essential democratic heritage.That is their view of real American history, and so be it. But to cast it aspart of the search for the ultimate truth of the American historicalexperience, especially when such a search is undertaken and describedas part of a self-styled practical realist perspective, is clearly an ideo-logically freighted conception produced by their social, political, andintellectual agenda. This is not, of course, problematic if practical real-ism with, in this case, its multicultural objective is acknowledged to bejust another set of ideological positions as represented in their narra-tive. Their book title, Telling the Truth About History, suggests that theyare mustering an arbitrarily contextualised, constructed and histori-cised narrative which imposes a particular set of signifying relation-ships on the past. Appleby, Hunt and Jacob do in fact recognise theirpresentist agenda in their attack on deconstructionist history with theirclaim that it is ‘The legacy of Cold War science . . . [that] helps explainthe cynicism, even nihilism, and certainly the intellectual relativism,

182 Conclusion

that greet even the mention of truth and objectivity.’9 So for them itseems that empirically derived truth remains transcendental, while thedeconstructionist questioning of its truth-value and reality-e!ect is epi-stemic and presumably transient, a product in this case of the psycho-logically destabilising e!ects of the Cold War. That this may or may notbe unfair is one thing, but what is important is that we always bear inmind that they are telling us their version of the truth about history.

The historicisation of Telling the Truth About History is a clearexample of Foucault’s argument that the past imagined as history existsonly in the contemporary discourses of historians. History remains aconstruction, whether it is viewed as a rhetorical representation or asempirically tested social theorising. Books with a claim to historicaltruth, like Telling the Truth About History, confirm my doubts aboutanyone’s ability to interpret and contextualise evidence as if in anideology-free ‘clean room’. As I have attempted to demonstrate in thisbook, the historian’s dialogue with his/her evidence cannot be under-taken through an objective, non-intertextual, non-figurative and value-free medium. As David Harlan has suggested, this leaves us with asmany di!erent kinds of histories and methods as there are kinds ofhistorical writing. And, as Foucault and White have suggested, thereasons for constructing the form of history in a particular way areusually ideologically inspired.

As Foucault argued in his genealogical analysis of the treatment ofthe insane and the exercise of power over the human body, history is acentral discourse instancing and ratifying the exercise of power. His-tory’s authority is at its most potent when, in the hands of disinterestedhistorians, it professes to reveal the objective truth of the past as itactually was. What such history does in pursuit of this assertion isrepresent a version of the past through what White called the shapingof language and cultural self-interest. How the evidence is translatedor narrated into the historian’s facts is basic to the exercise of power –what White has designated the ‘terminological determinism’ of thefigures of speech.10 This fundamental inability of historians to get tothe truth of the evidence is not, indeed, a debilitating experience,instead it permits a space to open up that cannot be so easily colon-ised for ideological purposes. Uncertainty in history is a form of pro-tection against what is politically right, wrong, correct or incorrect.Political certainty is, I would suggest, always suspect. The deconstruc-tionive consciousness position that history is not a summary of thetruth that emerges from the evidence will upset only a few particularlynaïve historians. But few in the mainstream would accept that whenwe write history we are creating a verbal/textual fictive artifact

Conclusion 183

generating what I have called a historical truth – the reality or truth-e!ect. They would not accept that the reality of the past resides in thetruth-e!ect or plausibility of the stories they tell as history, but wouldinsist that the evidence must remain the ultimate and absolute measureof the facts.

THEORIES OF HISTORY: CONSTRUCTING THE PAST

On the question of the construction of social (or rhetorical) frame-works with which to interrogate the evidence, the deconstructive con-sciousness recognises this as the impositionalism of the historian. Thesuspicions of naïve empiricists about critical approaches to historicalunderstanding are amply demonstrated in their antipathy towardsidealist and relativist philosophers of history like Collingwood, Croce,Beard, Becker and E.H. Carr, as well as innumerable deterministicsocial, cultural and Marxist-inspired constructionist historians, butmost recently to narrativists ranging from Michel Foucault, HaydenWhite and Dominick LaCapra to Louis Mink and Frank Ankersmit.The common link between these relativist historians is their assumptionnot only that they are in an interrogative dialogue with evidence butthat they are directly interventionist. In their di!erent ways, all recog-nise that the past becomes history only when it is construed through thefilter of the historian’s strategies of explanation. As a general position,mainstream historians accept the relativism inherent in construction-ism, short of taking it to its deconstructionist conclusion that e!ectivelyabandons the foundation of empiricism.

Constructionists understand past events through a variety ofmethods, econometric and statistical, employing anthropological andsociological deductive–inductive generalisations as covering laws, evenCollingwood’s empathic rethinking of the past. While for reconstruc-tionists the evidence o!ers up the truth through the examination of itsfine detail, for positivists it is prised out through the leverage ofappropriate theory. For the mainstream, these two extremes have beencompromised by E.H. Carr’s popular but, from the deconstructive per-spective, unconvincing view that while it is historians who write historyand create models of explanation, they do so according to the dictatesof the evidence. Carr’s method means that the continuous process ofrapid shifting between text and context is always informed by thestructures and patterns, o!ered up as the theories/models/concepts ofclass, race, gender, and so forth, that are to be found/discovered in theevidence. For E.H. Carr the evidence suggests certain appropriateexplanatory models of human behaviour which will then allow for ever

184 Conclusion

more truthful historical explanation. Most historians in the mainstreamwould likely accept this description of what they do.

Because, however, the demarcation between where empiricism endsand hypothesis testing begins is usually too hard to calculate, andincreasingly we accept as reasonable that historians are always activein the creation of the past through model-building, then why is itunreasonable to take into account our prefigured narrative emplot-ments, arguments and their ideological implications? Although hard-core reconstructionists in particular always maintain that moraldecisions have no place in the objective reconstruction of the past, Iwould contend that because no history is written that is free from pre-figured emplotment and argument, or moral and ethical positioning, afuller understanding of the past can emerge only when the imposition-alist role of the historian is fully appreciated as being that of an authorrather than that of a reporter.

Empirically founded constructionism today encompasses many newmodes of analysis and conceptualisations acquired from cognate discip-lines like sociology and anthropology. John Tosh’s description of his-tory as the product of hypotheses to be ‘tested against the evidence’attests to the complexities of mainstream history today. But history hasburst the banks of the mainstream as many practitioners try to bridgefact and fiction by borrowing analytical techniques from literary criticaltheory and by viewing history as a literary genre. Although cast asworries, the comments of Lawrence Stone are more than adequate evi-dence for this development. Rather than a cause for concern, it seems tome that historians should move in this direction, especially givenPhilippe Carrard’s recognition that not even the most positivist-inspired constructionist historians can escape the power of narrativeand emplotment – writing without tropes is hard to do.11

As I have tried to explain, the emphasis on the constitutive role of thehistorian’s narrative derives from our insights into the nature of lan-guage and the arbitrary nature of the sign. Historical narrative dependsmore on rhetorical rather than logical argument, working as it does bycollapsing the link between signifier and referent to produce the illusionof language transparently representing or corresponding with theevents of the past, and thus showing them in their ‘true light’. Themodels that we construct are similarly produced. It is at this point thatHayden White’s tropically prefigured model of historical narrativeassumes its significance. Once we do away with the notion of theunclouded connection of the word and the world, and view it asthe trope of realism, we can start to appreciate the importance of thenarrative strategies that White suggests enable historical interpretation.

Conclusion 185

Where deconstructionism parts company with constructionism is inthe latter’s positivist insistence that it is possible to relate and under-stand discrete events only by reference to a grand narrative of explan-ation like a covering law, or generalisations about human behaviourthat generally hold in given material situations or contexts. In thissense, constructionist a priorism elicits as little conviction in thedeconstructionist mind as it does in that of the reconstructionist.Constructionists are e!ectively claiming that truth emerges morerealistically through their particular empirically derived social theorythan it does through the reconstructionist recovery of fine detail, or, forthat matter, through the deconstructionist preoccupation with thetruth- or reality-e!ects of narratives. In spite of his early flirtation withthe Annales school, Foucault’s intention to undermine logocentrismwas, I have argued, eventually exemplified in his model of the rhetoric-ally and socially constructed episteme. It was his aim to locate themeans by which knowledge is produced linguistically within society. Hissearch turned him towards the historical rules of social change. Indoing this, unlike most historians who view change over time as thediscovery of the linear or diachronic unfolding of a coherent narrativeor interconnected process, Foucault views it as a network or synchronicstructure of power relationships, the aim of which is to create knowl-edge, and in particular historical knowledge. As he insists, history isnot about factual discovery but about the literary and textual creationof knowledge for the purpose of the exercise of power, or to countersuch exercise as a form of literary dissent. In this sense we historians,like the literary critic, or any overtly committed intellectual, areentangled in the webs of signification that we and society have created –historical knowledge is always implicated in discourse and culture.

Foucault, like White, draws on the tradition of Vico and assumes thathistorical change results from the interaction of human consciousnesswith its social and natural context, and this makes literary and culturalartifacts, such as our written history, little more than the hindsightrationalisations of human beings interacting in social situations. Thisoriginal insight of Vico, built on the incapacity of humanity ever tounderstand fully the natural world as well as we can know our ownsocial creations, has inspired much social science theorising. But thepositivism of the majority of constructionist historians, particularlystage theorists like Marxists, e!ectively ignores Vico’s insight that his-tory, as a literary art, necessitates a distinctive conceptual approach tothe analysis of past social and human phenomena, quite di!erentto the deductivism that characterises the study of their natural orsocial worlds.

186 Conclusion

Like Vico, White and Foucault both accept the creative power oflanguage. The complex manner in which we use language and languageuses us to mediate past reality suggests that no amount of sophisticatedsocial science hypothesis-testing can avoid the interactive relationshipbetween the historian, the word and the world. Narrative is not simplythe representation of the world of past reality, a reproduction of thingsand the relations subsisting between them. While language is used bymainstream historians as if it were capable of reproduction, it is primar-ily an innovative medium that has the power to invent and create ourknowledge of the past. Both White and Foucault, like Vico beforethem, have found this explanatory power of narrative in its metaphoricor troping nature.

NARRATIVE

I have argued in favour of narrative as history’s primary cognitivedevice working in the mind of the historian as he/she imagines, shapesand represents the past. The emplotment of history as a story, with itssupporting arguments and ethical strategies of explanation, is a highlycomplex form of explanation of historical change, but it is not historyas it actually happened. How the past is configured hangs on the histor-ian’s capacity to match a type of emplotment with the historical eventsthat he/she wishes to furnish with a meaning of a particular kind. Nar-rative explanation is more than recording a flow of events in their orderof occurrence. Lemon’s definition, ‘this happened, and then that’, sig-nals the sophisticated level of interpretation that is rapidly entailedbeyond mere sequencing. The historian’s function, noted by Gallie asthe essence of historical understanding, is to o!er a story that is follow-able. Such followability emerges when the stories that historians tell arecoherent and appear plausible in the light of the available evidence. Thereality of the past does not already exist in the unhewn marble, requir-ing only the historian’s skill to chip away to reveal the object existingwithin. This is certainly White’s position, but we can again pose thequestion: is there a narrative in the past to be retold?

Even though Collingwood and E.H. Carr admitted the existence of aconstant interplay between the historian and the events described, theywere still ultimately unwilling to accept that the resulting history wasprimarily a fictive enterprise. For them and others, more recently Rich-ard T. Vann, White’s argument that history is a tropically prefiguredliterary artifact, a story as much invented as found, remains unaccept-able because without the anchor of a determining meaning to be dis-covered in the evidence, facts cannot emerge and there is no standard

Conclusion 187

against which the truth of those facts can be measured. They take toheart Louis Mink’s comment ‘if alternative emplotments are basedonly on preference for one poetic trope rather than another, then noway remains for comparing one narrative structure with another inrespect of their truth claims as narratives’.12 But accepting Mink’sargument must not blind us to the problematic nature of our enterprise.The fundamental empiricist tenet that the truth is ‘out there’ remainsthe basic flaw in our understanding of what it is we do, and how we doit. History as a written discourse is not indi!erent to the forces thatcreate the past. Because the past is irrecoverable, its secondary writtenrecord is at least as important to our historical understanding as theevidence of the past itself. Our only access to the past is throughthe imaginative narrative and the intellectual operations registered inthe tropes, the theory of which provides us with a radical andpromising basis for categorising the historical imagination in any givenepisteme.13

The human troping capacity is complex. I have indicated how tropesoperate in the imaginations of historians by reductions and integrationsto represent the nature of change in the past. The important point ofFoucault’s historical method is the way in which he can be seen astaking the tropological foundation to human consciousness as a modelthrough which to evaluate how history emerges from the exchangebetween reality and language, or, as White describes the relationship,between ‘transitions in societies and the tropological transformationsof speech’.14 In this fashion it seems reasonable to me to argue thatnarrative, as the linguistic medium of human consciousness, may facili-tate historical change over time, not merely our description of it. Takentogether, Foucault and White o!er the figurative process as a modelthrough which historical change is lived and can be explained.

I have argued that because it is the tropological model of narrativethat frames the interpretation of the evidence, rather than the other wayround, historical understanding is as much the product of literaryartifice as it is a knowable historical reality. The rejection of the cor-respondence theory does not mean that we are completely free to selectany tropic–emplotment–argument–ideological configuration for theevidence, and then proceed to some ultimate historical version of liter-ary deconstruction that allows any meaning to be imposed on the pastwhile declaiming any responsibility for it. What we have instead is arecognition that there is a substantial degree of reciprocity between themental prefigurative process and the evidence, in as much as each nar-rativised piece of evidence is already an intertext that has previouslybeen interpreted and textualised by other historians working within the

188 Conclusion

archive and their episteme. No historian can work in ignorance of pre-vious interpretations or emplotments of the archive.

Naïve realism does not take su"cient notice of the great power oflanguage to describe and invent. Empiricism necessarily sells historyshort. As White says, language ‘used to describe a field of historicaloccurrences in e!ect constitutes the field itself ’ and, therefore, limitsthe kind of methods that can be used ‘to analyse the events occurringwithin the field’.15 That this great power of language resides in itsmetaphoric and figurative (tropological) structure can be understoodthrough an illustration. Comparing Theodore Roosevelt, upon hisreturn from big game hunting in Africa in 1910, with Halley’s Comet isa legitimate historical interpretation, from which we infer that Roose-velt was akin to a spectacular force of nature (technically this is asynecdoche used to signify Roosevelt’s quality of character). But, whilethis description explains and interprets Roosevelt, it in no way ties thelanguage that the historian uses to the events discussed.16 There is nonatural correspondence between Theodore Roosevelt and a comet. Inthe course of employing this image, the historian vividly paints a pic-ture of Roosevelt that brings him to life and places him within hiscontext (Halley’s Comet appeared in 1910). It also evidences White’sview that the historical narrative ‘does not image the thing it indicates; itcalls to mind images of the things it indicates’.17 In other words, themetaphor cannot constitute a genuine image of the object it aspires todescribe, and o!ers instead a cognitive map for the reader to find theappropriate (and explanatory) associative images. Comparing TeddyRoosevelt to Halley’s Comet is not referential nor less mimetic, but itremains meaningful because of its poetic character. The reality-e!ecttakes over – not to subvert reality but to create meaning.

The influence of deconstructionist history is seen today in the wideacceptance that the past, as written history, is a textual product of itsage and, given the central organising role of the historian, is inevitablyinflected by presentist ideological demands and the current dispensa-tions of power. It is increasingly accepted that the historian, throughhis/her narrative description, is fully implicated in any written represen-tation of pastness. Few see history as a matter of following the evidencelike footprints in the sands of time towards truth. More and morehistorians today feel much happier asking not just how did historicalactors understand their own lives and events which shaped them, buthow can those observer-historians build again their subjective world-views and explain their actions? Where the mainstream diverges fromdeconstructionism is not over the fact that history is concerned prim-arily with debates between narrative interpretations, but over the

Conclusion 189

deconstructionist insistence that objectivity is impossible to achieve.The mainstream will not take that deconstructionist step that o!ers‘Man’ (his actions, thoughts, behaviour, decisions) as being what his-tory is about; the historian inevitably and unavoidably must give it aform or shape by inventing, for example, the proletarian, the ‘de-natured woman’, ‘the origins of American Revivalism’, ‘a century ofwar’, ‘1921: the end of Soviet idealism’, the post-colonial ‘other’, thefirst industrial nation, the Third American Revolution, the Age ofEquipoise, the Uncle Tom race leader, or the father of the nation. Allthese are objects created by the historical imagination because itseemingly cannot capture and reproduce past realities. As Megillpoints out, no amount of empiricist flag-waving can deny that writtenhistory requires a discursive form of interpretation, what Ankersmitcalls the constitution of a linguistic object, and White the ‘poeticalunderstructure’ of written history.

CONCLUSION

So, to return to the question I posed at the outset: to what extent ishistory, as a discipline, the accurate recovery and representation of thecontent of the past, through its popular form of the narrative? Myanswer has been that as a vehicle for historical explanation theadequacy of its narrative structure must be judged within the widerpostmodern critique of the nature of meaning and language. The over-arching implication is that history can be no more, nor less, than arepresentation of pastness. Such a conception explicitly rejects writtenhistory primarily as an empirical discipline that purports objectively torepresent a presumed past historical reality. The issue is the nature ofrepresentation, not the empirical research process as such. The problemis to warn against the belief that we can truly know the reality of thepast through its textual representation. There is still a strong tendencyfor history in its narrative form to become more real than the reality,like America’s frontier experience represented through the frontierthesis of Frederick Jackson Turner. To Americans this history becameso important as a metaphor for American individualism and democ-racy that it took on an essential but wholly mythic dimension. As thehistory text becomes more real than the past itself, all the traditionalnotions of truth, referentiality and objectivity, which, paradoxically,gave rise to its status as historical truth, fade.

The past is not discovered or found. It is created and represented bythe historian as a text, which in turn is consumed by the reader.Traditional history is dependent for its power to explain like the statue

190 Conclusion

pre-existing in the marble, or the trompe l’æil principle. But this is notthe only history we can have. By exploring how we represent the rela-tionship between ourselves and the past we may see ourselves not asdetached observers of the past but, like Turner, participants in itscreation. The past is complicated and di"cult enough without theself-deception that the more we struggle with the evidence the closer weget to the past. The idea of the truth being rediscovered in the evidenceis a nineteenth-century modernist conception and it has no place incontemporary writing about the past.

Conclusion 191

Glossary

Aesthetic turn This describes the sensibility of historians to the aes-thetic (representational, poetic and literary) nature of history. It followson from the linguistic turn with its new emphasis upon the nature of thehistory text as a representation rather than simply as a linguistic con-struct. The issues of mimesis (mimicking the past through its represen-tation) and substitution of word for world are central. As Frank R.Ankersmit has maintained historians should understand the criticalaesthetic nature of history in which rational thought and empiricismare embedded. He argues the aesthetic decisions of the historian pre-cede cognition because history and not the past is where they start. Thebest-known reply to the aesthetic turn was o!ered, actually in advanceof the term being used, by the philosopher of history Peter Gay, in hisStyle in History (1974), which was a study of the rhetorics of Gibbon,Ranke, Macaulay and Burckhardt. Gay maintains that the historian’sliterary devices simply serve reality.

Ankersmit, Frank R. (1945–) Frank R. Ankersmit advocates a phil-osophy of history that views the undertaking as a narrative-makingactivity. He is a prolific writer, who has been published in several lan-guages. His first substantial text was Narrative Logic: A SemanticAnalysis of the Historian’s Language (1983), which was followed by acollection of his major journal articles, History and Tropology: The Riseand Fall of Metaphor (1994), and most recently three books on thenature of representation: Historical Representation (2001), PoliticalRepresentation (2002) and Sublime Historical Experience (2005). Inthese texts Ankersmit has provided, arguably, the most satisfactory andlucid account of the nature of the historical enterprise viewed as a formof narrative explanation. Essentially, he argues, the cognitive functionof history is located in the narrative substance of the text rather than itssingle statements of justified belief (factual statements). It follows that

history (i.e. a constructed narrative about the past) cannot be comparedwith the past itself. Only narratives can be compared with narratives.This judgement entails a fundamental reassessment of what is history,namely that empiricism is not the base unit of history and, con-sequently, we can only ‘know’ the past through its representation. Onlydescriptions can, therefore, be true or false.

A priori knowledge A term popular in philosophy that assumesknowledge to be independent of experience.

Argument A set of premises and the conclusion drawn or inferredfrom them. An argument is said to be valid (not the same as true) if theconclusion follows either inductively or deductively from the premise(s).

Cliometrics The application of mathematics and statistics to the dataof the past in order to facilitate interpretation. Popular in the 1960s,especially in economic history, but somewhat in decline in the 1990s.

Colligation In history, the process of explaining an event by bringingtogether a set of otherwise apparently separate events under a generaldescription or principle, e.g. the inventions of the eighteenth and nine-teenth centuries under the description of the revolution in scientificthought. The process is similar, but not identical, to emplotment whichidentifies a pattern in the sequence of events. For historians who rejectthe notion of emplotment in favour of colligation, its outcome is usu-ally assumed to be the more or less accurate reconstruction of the pastas history, because of the identification of causes. The danger is thathistorians may be tempted to be too tidy in relating events, althoughreconstructionist historians would argue that creating meaning throughthe context avoids such di"culties.

Constructionism The requirement of historians to propose rather thandiscover relationships between events in the past. In the twentieth cen-tury constructionism is most clearly found in the Marxist school whicho!ers the construct of class exploitation as the model for historicalunderstanding. The Annales school also produced a constructionist his-tory that suggested alternative demographic and behavioural theories.Constructionist historians are to be clearly distinguished from the othertwo major categories of reconstructionists and deconstructionists.

Context In history, the background to the events described, knowl-edge of which assists in the creation of meaning. In practice the

Glossary 193

context is the framework of other facts, events and precedingcircumstances.

Correspondence theory of truth The argument that propositions aretrue when they correspond with the facts. Although it is often di"cultto establish what is factual, the commonsense idea of a correspondence(or reflection) between the word and world remains for many historiansin the reconstructionist mainstream an attractive, if increasinglydisputed, notion.

Covering laws A model for historical explanation (directly related toestablishing causes) developed by the American philosopher of historyCarl Hempel (1905–), and founded on the insistence that an event maybe explained when it is capable of being deduced from a law of natureor human behaviour. Very often its form is that of statisticalprobability.

Death of the author Derived from the deconstructive study ofliterature originating with Roland Barthes (1915–1980) and employedextensively by Michel Foucault (1926–1984) that suggests that all textsprecede their authors who are simply constructs who cannot privilegemeaning. For the deconstructionist historian evidence does not denote adiscoverable past reality as found in author intentionality, but o!ersinstead only chains of significations and interpretations.

Deconstructionism A term originating with Jacques Derrida (1930–2004) that suggests that understanding texts is not solely or exclusivelydependent upon reference to the external reality of empiricism, God,reason, morality, objectivity or author intentionality (see Death of theauthor). This logocentric notion of an originating source of absolutemeaning is disputed in favour of the assumption that meaning isarbitrary and figuratively produced.

Deconstructionist history In history, a model of study that questionsthe traditional assumptions of empiricism couched as factualism,disinterested analysis, objectivity, truth, and the continuing divisionbetween history, ideology, fiction and perspective. Instead, decon-structionist history accepts that language constitutes history’s contentas well as the concepts and categories deployed to order and explainhistorical evidence through our linguistic power of figuration.

Determinism The notion that historical processes are structured

194 Glossary

according to forces beyond individual or group choice or influence, sothat in e!ect all events are unexceptionably e!ects determined by priorevents. The most famous example is the Marxian model which insiststhat history is the result of the class struggle.

Di!erence/di!érance Term coined by deconstructionist philosopherJacques Derrida (1930–2004) as a play on the French verb ‘di!érer’which means to di!er and defer. The consequent slide of meaning asbetween signs means that we cannot distinguish a true or original pointof meaning – there is no original site of meaning. A central argument inpost-structuralism.

Discourse The result of the placement or insertion of a text (usuallylonger than a simple sentence) into its context so that it derives a coher-ent meaning shared by both author(s) and reader(s). As a shared lan-guage terrain, a discourse has reference to extra-linguistic dimensionsas found in the material and ideological worlds of institutional andeconomic power.

Empathy Usually associated with the historical method espoused bythe British historian R.G. Collingwood (1889–1943) in The Idea ofHistory (1936), it means the state of being ‘in touch with’ the thoughtsand situation of the historical agent. The route to this emotional andintellectual state is to interpret the historical evidence by literallyrethinking the thoughts of people in the past within their known con-text. For many reconstructionist historians it is the empathic linkage ofevidence and context that constitutes history. See also hermeneutics.

Empiricism The method whereby knowledge is gained through the useof the senses as we observe and experience life, or through statements orarguments demonstrated to be true. In the Anglo-American tradition ofwriting reconstructionist history, empiricism has been the central meth-odology with a particular insistence on the corollary of the objectiveobservation of the reality to be discovered ‘out there’. The problemusually encountered by empiricism is that thought does not simplyemerge from experience, but actually provides us with concepts or men-tal categories that we utilise to organise and make sense of our experi-ence. This inevitably leads to the question: how can we truly know thereality ‘out there’, given that our observations may well be only con-structions of our mind or intuition? Most empiricists and moderate orpractical realist reconstructionists today accept a middle position, thatwe observe but we also mentally process information, deploying a priori

Glossary 195

knowledge as appropriate and helpful. Empiricism can, of course, takethe form of a denial of a priori knowledge.

Emplotment The meaning of any historical or more overtly fictionalnarrative is provided by the emplotment (story line or plot structure),i.e. a narrative of events and their causal, contextual or colligatory con-nections. It is the role of the historian to turn a sequence of events (thishappened, then that) into a story of a particular kind – romantic,comic, tragic or satiric, or some combination. Dependent upon themethodological inclination of the historian, the emplotment is pro-duced with the intention of discovering the meaning or imposing ameaning on the events. All histories have emplotments.

Enlightenment (the) A widespread intellectual, cultural and techno-logical/scientific movement, being the seedplot of the Modern Era. Itbegan in the early seventeenth century in England (preceded by thework of René Descartes, Francis Bacon, John Locke and ThomasHobbes) and ended at the close of the eighteenth in France and Ger-many (with Voltaire, Diderot and Lessing), but was found throughoutEurope. European thought was characterised in what was a time ofgreat technological and scientific change by an acceptance of newconcepts like positivism and experimentation in science, by the closeobservation of natural phenomena, reason and rationality promotingexplanation, by new ideas concerning government through contractrather than force (based on the emerging doctrine of Liberalism with itscentral tenets of popular sovereignty and equality of opportunity), andby a new conception of the market-place as a rational economic mech-anism. Its impact on history is seen in its very creation as a disciplinefounded on the belief that it was the record of progress and humanperfectibility. Perhaps inevitably the thought it generated eventuallyturned in on itself, promoting a questioning of its own central tenets insucceeding centuries, notably in the present (or postmodern age).

Episteme Michel Foucault (1926–1984) in The Order of Things (1966)uses the term to designate how a culture acquires and organises knowl-edge in a given historical period. The episteme connects all the separatediscourses (religious, scientific, historical, medical, etc.) into a more orless coherent structure of thought founded on a set of shared assump-tions about how such knowledge is obtained and deployed. The sharedassumptions are fixed through the troping process (trope/figuration)which takes place at the deep level of the human consciousness, andwhich is basic to the emplotments that historians generate. Knowledge is

196 Glossary

thus organised within each of the four distinctive historical epistemes,which Foucault maintains existed from the sixteenth to the twentiethcenturies. For the historian these assumptions or attitudes, as theycharacterise each age’s dominant form of narrative representation, aredisplayed in our narratives and directly influence our access to the ‘real-ity’ in the evidence through our metaphoric encodation of similarity ordissimilarity.

Epistemology The theory of knowledge. Among its concerns are howit is discovered/constructed through various mechanisms/methods likeempiricism or a priori knowledge, how we can justify what we believe (asopposed to what we know), and scepticism in knowledge acquisition. Itremains one of the central concerns of philosophers, and is increasinglya growing interest among historians because of the disputed role ofempiricism as the historical method.

Ethical turn The idea of history as an ethical pursuit has gained muchground. In terms of both the use to which history can be put and how itis constructed, ethics plays a central role. Arguably, history is just oneamong all those discourses that are secondary to ethics. Importantly,this casts doubt on the notion that we can learn moral (or any) lessonsfrom the past. Not only is the present di!erent to the past and almostcertain to be di!erent to the future, the aesthetic turn suggests that howwe represent has built into it an ethical value. History, like any con-structed narrative, is secondary to ethics, which the moral philosopherEmmanuel Levinas (1905–1995) called a first philosophy. Arguably, his-tory is post-ethical and the appeal of any history lies in representingthose political and moral principles that most appeal to us. So, if his-tory is unavoidably an ethical activity, historians must recognise themoral preferences they have. The ethical turn suggests moral lessons donot materialise from the past; they are embedded in history as weconstruct it.

Evidence Traditionally, the sources, both documentary (primary) orwritten by historians (secondary), upon which authoritative historicalexplanations are founded. Evidence cannot be considered as separatefrom the process of its interpretation through inference, its constitutionas fact(s) by an initial verification and comparison attesting to itsauthenticity, and by being set within its context.

Fact The concept of the fact is a complex and contentious one amonghistorians. Traditionally, a fact is the actual and undisputed event,

Glossary 197

process or piece of social action upon which historians agree – theBattle of Waterloo occurred in 1815 – the correspondence of reality anddescription. Beyond this simple level of factual statement historiansimmediately enter the realm of interpretation. What do we do with thefact(s)? How do we colligate them? How do we emplot/narrate them?How do we sequence and explain them? Beyond the usual problemswith the evidence – that it may be of doubtful authenticity, or unreliable(their authors lied (!)), or simply absent – historians have many di"cul-ties in constituting facts. What criteria should be used by the imposi-tionalist historian to winnow out that evidence judged to be irrelevant tothe constitution of a fact? How reliable is inference as a method forestablishing facts? Should historians all become constructionists ‘test-ing’ evidence against a hypothesis to establish a fact? What of theunreliable nature of the signifier–signified–sign equation?

Hermeneutics Literally the art of interpretation of texts (evidence). Atechnical skill used by post-Reformation Protestants to interpret theBible, modern hermeneutics began with the e!orts of FriedrichSchleirmacher (1768–1834) to understand texts grammatically andwhat the authors are likely to have intended when writing them (gram-matical and psychological elements). The hermeneutic circle is the loopbetween the text and the author. As a process of interpretation it wasextended by Wilhelm Dilthey (1833–1911) to include drawing analogiesbetween the likely intentions of the author of the evidence and our ownexperiences. This formed the basis of R.G. Collingwood’s (1889–1943)notion of empathy. In the twentieth century Martin Heidegger (1889–1976) further extended hermeneutics to include the interpretation ofour own being and existence as interpreters.

Historical interpretation A narrative account of the events, occur-rences, texts and people in the past that makes the content understand-able and/or plausible. The process at some point involves all aspects ofscrutiny of the evidence and historical method, how so ever it is definedby the historian – inference, colligation, contextualisation, emplotment,argument, impositionalism, empathy, etc.

Ideology A coherent set of socially produced ideas that lend or createa group consciousness. Ideology is time and place specific. Constitutedas a dominant mode of explanation and rationalisation, ideology mustsaturate society and be transmitted by various social and institutionalmechanisms like the media, Church, education and the law. In the viewof some commentators, ideology is to be found in all social artifacts like

198 Glossary

narrative structures (including written history), codes of behaviourand patterns of belief. Ideology, according to Marxian theory, reflectsand maintains the authority of the dominant social class by delib-erately obscuring the reality of economic exploitation, thus ensuringthat the economic relations of capitalist society appear natural andlegitimate.

Impositionalism The process whereby historians are implicated in theconstitution of the past as history. Although rejected by naïve realists asa corruption of the historical enterprise, practical realists and decon-structionist historians recognise the unavoidable nature of the dialoguebetween the historian and his/her sources. The latter group, in particu-lar, accept that historical interpretation means arranging ideas, sortingevidence, and imposing an explanatory emplotment or argument on thepast. It follows that historical knowledge is produced as a linguistic textthat has no direct access to past reality.

Induction/deduction Induction is a form of explanation based uponinference from the particular to the general, or it may be describedalternatively as generalising from observed instances. It is the trad-itional or common form of historical explanation. Deduction is explan-ation where a conclusion must follow logically from a set of premises. Inpractice, most historians employ both methods in explanation.

Inference The thought process of moving from one set of beliefs toanother based upon fresh information. Its two primary forms areinduction and deduction.

Jenkins, Keith (1943–) Keith Jenkins is the leading British historicalsceptic who, from the early 1990s with his (in)famous text RethinkingHistory (1991), confronted the flaw that he argues exists at the heart ofthe history enterprise. This is the manner in which the discipline hidesits true nature as a representation from itself (and, therefore, from itsconsumers). Viewing himself as an intellectual outsider, he made thecontroversial judgement that history was and is a cultural, literary andphilosophical activity that produces meaning about the past rather thandiscovering it in its empirical traces. His analysis followed through theconsequences of seeing history and the past as ontologically distinctcategories. The one is what once was and is now gone and is unrecover-able; the other is a discourse about it. In essence, it has been his argu-ment, in several important books, that the epistemological fixationof historians has been at the expense of understanding the nature of

Glossary 199

history as a text. Until his book Why History? (1999), Jenkins hadbroadly been acknowledged as a postmodern historian who still wishedto engage with the past through history. But, with Why History? hechanged his orientation, arguing that we should now forget history andlive according to the ideas o!ered by other epistemologically scepticaltheorists like Roland Barthes, Michel Foucault, Jacques Derrida, Jean-François Lyotard, Jean Baudrillard, etc. He is now committed, espe-cially since Refiguring History (2003), to existing in and theorisingabout a world with no history. It is from this position that the aestheticturn has emerged.

Linguistic turn An umbrella term describing a number of strands inWestern thought in the twentieth century, but especially found in post-structuralist thought, stressing that the route to knowledge invariablycentres upon the role of aesthetics, discourse and forms of representa-tion in and through language. The linguistic turn centres on the opacityand figurative character of language, the manner in which subjectpositions as well as reality-e!ects are created within language.

Logocentrism Term used frequently by Jacques Derrida (1930–2004)and advocates of deconstructionism in history, literature and philosophyto criticise the idea that there can be any fixed or centre to meaningestablished independently of language, and/or that language (especiallythe spoken word) can authentically represent reality.

Meta-narrative Literally a narrative about narratives, the term wasused by Jean-François Lyotard (1924–) in his book The PostmodernCondition: A Report on Knowledge (1984) in which he argued that meta-narratives or master-narratives, the stories told about how we gainedknowledge and thus understood human progress and history (Hegelian-ism, Marxism, Liberalism, the Enlightenment), have reached the end oftheir useful life in what is now the postmodern era. The fact that we canno longer depend on such grand stories as universal bench-marksagainst which to measure or ensure truth characterises our postmoderncondition. What we are left with are numerous ‘little narratives’ thate!ectively become self-legitimating.

Modernism/modernist Historically, modernism describes the nine-teenth- and twentieth-century movement in the arts, culture andliterature which in general terms criticised the positivist, objectivist,rationalist, empiricist and referentialist certainties of the Enlightenment.Confusingly, from the perspective of philosophy, modernism begins

200 Glossary

with René Descartes’ (1596–1650) search for rationality in understand-ing, and is thus regarded as being co-terminous with the seventeenth-and eighteenth-century Enlightenment. Michel Foucault (1926–1984)conceives the Modern episteme as constituting an epistemologicalincongruity for humanity, with Man as both the product of his socialexperience while also being the constitutor of knowledge throughdeduction.

Narrative A structure of explanation used to account for the occur-rence of events and human actions. At its most basic the historicalnarrative is the vehicle for colligation because it explains how thingshappen, and in what order, according to the cause and e!ect of ‘thishappened, then that’. When the historical narrative is constructedaround a selected emplotment it becomes the primary vehicle for thetransmission, and arguably the constitution, of historical understand-ing. What is often disputed is the extent to which the historical narrativecan correspond to the past as it actually was – that is, be able to recountthe story.

New empiricism This is the term used increasingly to describe the‘re-turn’ to an emphasis upon the knowability of the past in a post-representationalist age. Inevitably, given the substantial critique ofhistory as an exclusively empirical and analytical activity, increasingnumbers of constructionist historians have tried to recoup the centreground of historical thinking and practice by an attempted marriage ofaesthetic and linguistic self-consciousness with the desire to re-insinuateempiricism at the heart of the historical enterprise. Never abandoning abelief in objective reality and its ultimate knowability, new empiricistshave tried of late to recover agent intentionality as the key principlewhereby they can return to some form of chastened reality. In anyevent, it seems unlikely that such moves would satisfy those who occupythe farthest ends of the debate.

New historicism The revival of interest since the early 1980s in thestudy of literary texts within their historical contexts. New historicismis important to the writing of history because it draws upon that bodyof post-structuralist literary criticism that doubts the solidity of lan-guage as a clear medium capable of adequately representing the pastmaterial world. It also acknowledges that the historical text is generatedintertextually within the wider social and institutional context, and that,as a consequence, there are no absolute or transcendent truths to bediscovered, neither are there theories of explanation to be verified

Glossary 201

through empirical testing. The practical distinctions between factualand fictional texts are thus placed under doubt.

Positivism A theory of knowledge developed by the French sociol-ogist Auguste Comte (1798–1857) as part of his grand theory ofprogressive historical evolution over three stages, beginning with thetheological, then the metaphysical, and ending with the scientific orpositive stage. The final stage (which Comte saw himself as ushering in)is characterised by the verifiable or empirical measurement and predict-ability of the relationship between discrete phenomena. As an extensionof established notions of empiricism, positivism insists on no specula-tion about natural phenomena. Because positivism assumes a uniform-ity in scientific method, it allows for the analytical study of humanbehaviour – scientific sociology. Positivism’s legacy for the historian isplainly seen in crude forms of constructionism whereby historians mustaggregate evidence from which our forensic skills will generate incontro-vertible facts that operate according to laws of human behaviour (cf.covering laws). The historian is presumed to do this objectively withoutany impositionalism on his/her part.

Postmodernism A term used in many di!erent contexts (history,painting, literature, architecture, fashion, music) as a description ofthe various critiques of, and reactions against, the Enlightenment andits cultural product modernism. According to Jean-François Lyotard(1924–) in The Postmodern Condition: A Report on Knowledge (1984),postmodernism is characterised specifically by its rejection of overarch-ing grand or meta-narratives deployed in the modern historical era toexplain and justify human history and progress. The result is that thepostmodern age is distinguished by its post-structuralist-inspired denialof transcendental realities, fixed meanings, facts and the correspondencetheory of truth. Postmodernism as an approach to understanding thusproduces, among other things, tentative beliefs, playfulness, style andvogue, neo-pragmatism in philosophy, the linguistic turn, presentism,relativism, the reality-e!ect, deconstructionism and self-reflexivity in his-tory and literature, doubts about referentiality, and the ultimate failureof narrative as an adequate mode of representation. Postmodernismencourages doubt and uncertainty, challenges hierarchy and authority,and promotes the acceptance of ‘the other’ as legitimate.

Post-structuralism Claimed, as a part of the postmodernist move-ment, to be the successor to (and reactive against) structuralism. Theinspiration for the so-called linguistic turn in historical writing and

202 Glossary

understanding, post-structuralism insists that language, as the culturaland intellectual form, is the medium of exchange for power relation-ships (e.g. Michel Foucault [1926–1984] and power/knowledge) and theultimate constitutor of ‘truth’. Post-structuralism can trace its lineagethrough the work of various philosophers, historians and thinkers suchas Friedrich Nietzsche (1844–1900), Benedetto Croce (1866–1953),Martin Heidegger (1889–1976), Hans-Georg Gadamer (1900–),Jacques Derrida (1930–2004), Michel Foucault (1926–1984) and JuliaKristeva (1941–).

Reality-e!ect A concept explored at some length by Roland Barthes(1915–1980) in his essay ‘The Discourse of History’ (1967). Barthes’argument that the connection between language and history does notrely on any genuine conformity between evidence and its constitution ashistorical fact, means that what historians take for the past as it actuallywas, is only a reality-e!ect generated by our assumption that the cor-respondence theory of truth allows us to adequately reconstruct the past.As a result, the idea of historical truth becomes ever more problematicfor deconstructionist historians.

Reconstructionism One of the three major strands in historical inquiry(cf. constructionist and deconstructionist history). Reconstructionist his-torians range from conservative empiricists to practical realists depend-ent primarily upon their attitude towards the validity and practice ofcommonsense empiricism as the fundamental historical method. Moreprecise characterisation is complex, given its dependence upon the his-torians’ attitudes towards the uses of evidence, referentiality, etc., butparticularly upon how they envisage the role of language and narrativeas cognitive elements in the reconstruction of the past.

Referentiality A term used to designate a general belief in the largelyunproblematic or adequate match between reality (event, person, thing,process) and its description (linguistic expression). Structuralismteaches that words are not signifiers that relate in any natural fashion totheir referents – the things to which they refer in that the relationshipbetween the word and the world is arbitrary (socially provided) – and soit follows that any referentiality assumed in language is the resultof it being fixed in language by conventional usage. This situationcomplicates the translation of facts into interpretation in as much asreferentiality cannot be assumed to extend beyond the most basic level.

Relativism The idea that a precise measurement against a fixed

Glossary 203

benchmark is impossible in practice leads to the notion of uncertainty.In history, the relativist debate has been enjoined over many yearsbetween conservative empiricist- and/or positivist-inspired reconstruc-tionists and their ideal of a stable and objective history of the past, andthose who believe that the history they write is as much the product oftheir narrative and their present, as it is of past reality.

Representation Any sign, word, sentence, discourse, picture, sound oraction intended to depict or characterise another is an act of represen-tation. The correspondence theory of truth takes representation to becloser to reflectionism rather than, say, resemblance. For historians,representation is an important concept in that it forms the mechanismthat allows empiricism to work. There is a working assumption thatlanguage is, by and large, an adequate medium of representation for thereconstruction of the past. The empiricist foundation of history thusrepudiates the deconstructionist assumption that facts are literaryartifacts and, therefore, open like all texts to the post-structuralistcriticisms levelled at the taken-for-granted link between reality and itsrepresentation in language.

Signifier–signified–sign According to the structuralist model of lan-guage proposed by Ferdinand de Saussure (1857–1913), words are‘signs’ defined in their di!erentiation from other words, and notbecause of any natural link with the real world of objects/things. Signsare built of the signifier and the signified, with the word or conceptas the signifier and the thing represented as the signified = signifier–signified–sign. The arbitrary nature of the signifier–signified–signrelationship flows from its social or cultural constitution. Althoughhistorians constantly use words as if they were strictly referential, theyare based on invented meanings often derived from widely acceptedcultural values, and as Michel Foucault (1926–1984) argues, related toinstitutionalised power relationships within social structures. Thisinherent uncertainty in meaning prompted Jacques Derrida (1930–2004)to argue that it is impossible to write truthful narratives as historicalexplanations because there is no certain origin in linguistic meaning.

Structuralism A broad intellectual movement the high point of whichwas reached in France in the 1960s. Basic is the idea, derived from thework of Ferdinand de Saussure (1857–1913) in linguistics, that the rela-tionship between all discourses, cultural forms, belief and behavioursystems can be understood employing the structure of language as themodel. In practice, this means social meaning is generated according to

204 Glossary

the constrast between inherent binary opposites operationalised at thedeep level of human consciousness and revealed in the real world in thestructure of grammar, myths, sexual relationships, etc. For history thismeans its data is primarily understood through our linguistic mentalstructures rather than found in the external empirical data. Inevitablythis casts doubt on notion of evolutionary change, scientific objectivity,the disinterested search for truth, and referentiality. Structuralism’sdescent into post-structuralism has probably had a greater impact on thewriting of deconstructionist history.

Trope/figuration Taken as figures of speech (primarily metaphor,metonymy, synecdoche and irony, but we could also include the variantssimile, litotes, periphrasis and hyperbole) that deploy words in such away as to turn or translate meaning. Troping operates at the deep levelof human thought in Saussure’s sense of creating meaning throughbinary opposition, and as employed by Michel Foucault (1926–1984) inthe sense of otherness, or di!erence in any historical period. In his bookMetahistory (1973), Hayden White (1928–) examined the theory oftropes and troping as a means to distinguish the dominant modes ofhistorical imagination in nineteenth-century Europe. By extrapolationto the cultural level we may identify the deep and surface structuresof the historical imagination. The troping process may be extendedto include the creation of large-scale metaphors, like the base-superstructure metaphor of Marx as the basis of a total explanation ofhistorical change, or to create other models of historical change thatrely upon the basic relationships of part-whole/whole-part. The tropesmay thus be regarded as being at the heart of every historical period andin its description.

White, Hayden (1928–) Hayden White’s key works on history are tobe found in the path-breaking Metahistory (1973), Tropics of Discourse(1978), The Content of the Form (1987) and Figural Realism: Studies inthe Mimesis E!ect (1998). In these texts (and many others), White hasexamined the connection between what he refers to as the historicalimagination and the creation of the historical narrative. White is bestknown for his argument that history is as much the result of the histor-ical imagination and its written construction as it is discovery in therecords. Relatedly, he maintains that history does not match up to a pre-existing story. In other words, there is no built-in meaning to the past.Hence, it is the role of the historian to provide it. White insists weimpress our stories on the past for certain reasons that are essentiallyepistemological, and which will also be ethical and ideological.

Glossary 205

Moreover, the foundation for the logic of history exists in the power offiguration as it is with all forms of literature. So, the logic of history isnot primarily about empiricism and inference; it is, in fact, about itsconstruction as a literary artefact that entails the emplotment of thepast as a story of a particular kind.

206 Glossary

Guide to further reading

The notes and references indicate the sources and thinking behind myarguments and conclusions. This short guide is intended to signpostwhere you can turn for a more detailed study of the book’s key issues,about the character of history, the two methodological historical main-streams, and their deconstructive challenge. We began with history’straditional empirical or reconstructionist method. The basic principlesof this school are still nowhere better expounded than by G.R. Elton inThe Practice of History (London, Fontana, 1967) and his statement ofthe faith of the conservative empiricist in Return to Essentials (Cam-bridge, Cambridge University Press, 1991). A firm grounding in theclassic approach to reconstructing the past is, as always, o!ered byArthur Marwick in his The New Nature of History: Knowledge, Evi-dence, Language (Houndmills, Palgrave, 2001), which is justifiably oneof the most popular introductions to history imagined as a craft. Astout Marxist defence of ‘proper history’ is o!ered by Bryan Palmer inhis Descent into Discourse: The Reification of Language and the Writingof Social History (Philadelphia, Temple University Press, 1991). Mov-ing more towards the empiricist centre are the perennially popular JohnTosh, The Pursuit of History (London, Longman, third edition, 2001)and Peter Charles Ho!er and William W. Stueck, Reading and WritingAmerican History: An Introduction to the Historian’s Craft (2 vols, Lex-ington, D.C. Heath, 1994). The practical realist element of the moder-ate mainstream continues to be ably represented by Jerzy Topolski,‘Towards an Integrated Model of Historical Explanation’, History andTheory, Vol. 30, No. 3, 1991, pp. 324–338, and Joyce Appleby, LynnHunt and Margaret Jacob, Telling the Truth About History (New York,Norton, 1994).

Also very useful as a general introduction is the reader by AnnaGreen and Kathleen Troup, The Houses of History: A Critical Readerin Twentieth Century History and Theory (Manchester, Manchester

University Press, 1999). Still worth reading, though taken to be some-what more relativist in his approach to the creation of the past, is E.H.Carr, What is History? (London, Penguin, second edition 1987). Con-fronting the postmodern or deconstructive challenge to the classicparadigm remains Gertrude Himmelfarb, ‘Some Reflections on theNew History’, American Historical Review, Vol. 94, No. 3, June, 1989,pp. 661–670.

See also any of a substantial range of recent and very accessiblesurvey texts that include Ludmilla Jordanova, History in Practice (Lon-don, Arnold, 2000), Mary Fulbrook, Historical Theory (London,Routledge, 2002) and the collection by Peter Lambert and PhillippSchofield (eds), Making History: An Introduction to the History andPractices of a Discipline (London and New York, Routledge, 2004). Seealso the valuable and wide-ranging o!ering by Stefan Berger, HeikoFeldner and Kevin Passmore (eds), Writing History: Theory and Prac-tice (London, Arnold, 2003). The ‘new empiricist’ approach has beenextremely well served by the collection edited by Gabrielle Spiegel Prac-ticing History: New Directions in Historical Writing After the LinguisticTurn (New York and London, Routledge, 2005) and Carla Hesse’s art-icle, ‘The New Empiricism’, Cultural and Social History, Vol. I, No. 2,2004, pp. 201–208. There is, in addition, the introductory and surveyseries entitled Theory and History, edited by Donald MacRaild, whichincludes several texts on specific aspects of history. See, for example,Matt Perry, Marxism and History (Houndmills, Palgrave, 2002),Stephen Davies, Empiricism and History (Houndmills, Palgrave, 2003),the constructionist Willie Thompson’s Postmodernism and History(Houndmills, Palgrave, 2004), Donald MacRaild and Avram Taylor,Social Theory and Social History (Houndmills, Palgrave, 2004) andAlun Munslow, Narrative and History (Houndmills, Palgrave, forth-coming). A basic survey of the varieties of history is to be found inJ. Gardiner (ed.), What is History Today? (London, Humanities PressInternational, 1988) though somewhat more encyclopedic is MichaelBentley (ed.), Companion to Historiography (London and New York,Routledge, 1997). See, in addition, David Cannadine (ed.) What isHistory Now? (Houndmills, Palgrave, 2002), which o!ers an excellentcollection of fresh thoughts on the nature of the discipline, althoughthe general orientation is, unfortunately, sceptical of deconstructionistapproaches.

Defending the philosophical foundation of the reconstructionistapproach is still the province of its leading voice C. Behan McCullaghin Justifying Historical Descriptions (Cambridge, Cambridge UniversityPress, 1984) and more recently his The Logic of History: Putting Post-

208 Guide to further reading

modernism in Perspective (London and New York, Routledge, 2004).See, likewise, Chris Lorenz ‘Can histories be true? Narrativism, Positiv-ism, and the “Metaphorical Turn” ’, History and Theory, Vol. 37, No. 3,1998, 309–329 and his ‘Historical Knowledge and Historical Reality: APlea for “Historical Realism” ’, History and Theory, Vol. 33, No. 3,1994, pp. 297–327.

Possibly the best current general introductions to the methodologiesavailable in history today are Michael Stanford’s A Companion to theStudy of History (Oxford, Basil Blackwell, 1994) and his more recentAn Introduction to the Philosophy of History (Oxford, Blackwell Pub-lishers, 1998). Also of particular value is the reader edited by Robert M.Burns and Hugh Rayment-Pickard, Philosophies of History: FromEnlightenment to Postmodernity (Oxford, Blackwell Publishers, 2000).In addition, highly recommended is M.C. Lemon’s Philosophy of His-tory: A Guide for Students (London and New York, Routledge, 2003).The most comprehensive exploration of the modern development ofthe American history profession and its methodological concernsremains Peter Novick, That Noble Dream: The ‘Objectivity Question’and the American Historical Profession (Cambridge, Cambridge Uni-versity Press, 1988), but of singular importance is David Harlan, TheDegradation of American History (Chicago, Chicago UniversityPress, 1997).

The debate on historical objectivity is to be found in the AmericanHistorical Review Forum, ‘The Objectivity Question and the Future ofthe Historical Profession’, American Historical Review, Vol. 96, No. 3,June, 1991, pp. 675–708. More general philosophy of history textswhere issues such as objectivity, truth, and meaning are addressedinclude William H. Walsh, An Introduction to Philosophy of History(London, Hutchinson, third edition, 1967), William Dray (ed.),Philosophical Analysis and History (New York, Harper and Row, 1966),Leon Goldstein, Historical Knowing (Austin, University of Texas Press,1976) and Mark Bevir, The Logic of the History of Ideas (Cambridge,Cambridge University Press, 1999).

Although an increasing number of texts published from the mid-1990s to the late 2000s have tried to move ‘beyond’ the linguistic turn,early notable examples of the recognition of the turn are DominickLaCapra and Steven L. Kaplan, Modern European Intellectual History:Reappraisals and New Perspectives (Ithaca, Cornell University Press,1982), Lynn Hunt, The New Cultural History (Berkeley, University ofCalifornia Press, 1989). Also useful, though critical of Hayden White,is Saul Friedlander (ed.), Probing the Limits of Representation:Nazism and the ‘Final Solution’ (Cambridge, Massachusetts, Harvard

Guide to further reading 209

University Press, 1992). In the past decade or so, useful texts includeMichael S. Roth, The Ironist’s Cage: Memory, Trauma, and the Con-struction of History (New York, Columbia University Press, 1995),David D. Roberts, Nothing but History: Reconstruction and Extremityafter Metaphysics (Berkeley, University of California Press, 1995),Robert Berkhofer, Beyond the Great Story: History as Text and Dis-course (Cambridge, Massachusetts, Harvard University Press, 1995),Clayton Roberts, The Logic of Historical Explanation (University Park,Pennsylvania State University, 1996), Roger Chartier, On the Edge ofthe Cli!: History, Language and Practice (Baltimore and London,Johns Hopkins University Press, 1997), Georg Iggers Historiography inthe Twentieth Century (Middletown, Wesleyan University Press, 1997),Miguel A. Cabrera Postsocial History: An Introduction (Lanham,Lexington Books, 2004). D. Carr, T.R. Flynn and R.A. Makkreel, TheEthics of History (Evanston, Northwestern University Press, 2004) anda themed issue on ‘Unconventional History’ in History and Theory, Vol.41, 2002. An important voice in the debates on the utility of history inthe past decade has been Beverley Southgate, who has written a numberof important texts including History: What and Why (London,Routledge, 1996), Why Bother with History? (Harlow, Pearson, 2000),Postmodernism in History: Fear or Freedom? (New York and London,Routledge, 2003) and most recently What is History For? (London andNew York, Routledge, 2005). From 1997 to the present, the journalRethinking History: The Journal of Theory and Practice has been in theforefront of the ‘new wave’ of postmodern and experimental ways ofengaging with the past.

On the general relationship between postmodernism as an intel-lectual movement and the writing of history, see Stephen Bann, TheClothing of Clio: A Study of the Representation of History in Nine-teenth-Century Britain and France (Cambridge, Cambridge UniversityPress, 1984), Derek Attridge, Geo! Bennington and Robert Young(eds), Post-Structuralism and the Question of History (Cambridge,Cambridge University Press, 1987) and David Harvey, The Condition ofPostmodernity: An Enquiry into the Origins of Cultural Change (Oxford,Basil Blackwell, 1989). On the intellectual history of postmodernism,see Joyce Appleby et al. (eds), Knowledge and Postmodernism in Histor-ical Perspective (London, Routledge, 1996), which provides an excellentintroduction to the key texts among others of Nietzsche, Ricoeur,White, Foucault, Derrida and Rorty. The most recent and highly access-ible introduction to the issues confronting history from a postmodernperspective is that of Callum G. Brown, Postmodernism for Historians(Harlow, Pearson Longman, 2005). But also profitable is the recent text

210 Guide to further reading

by Alun Munslow, The New History (Harlow, Pearson Longman,2003). Cast in opposition mode is Ernst Breisach, On the Future ofHistory: the Postmodernist Challenge and its Aftermath (Chicago andLondon, University of Chicago Press, 2003). Illustrative of the avantgarde in historical thinking and practice is the collection edited andintroduced by Alun Munslow and Robert A. Rosenstone, Experimentsin Rethinking History (New York and London, Routledge, 2004) andKeith Jenkins and Alun Munslow, The Nature of History Reader(London and New York, Routledge, 2004).

The debate on social theory constructionism in history is still wellserved by Alex Callinicos, Theories and Narratives: Reflections on thePhilosophy of History (Oxford, Oxford University Press, 1995), Chris-topher Lloyd, The Structures of History (Oxford, Basil Blackwell,1993), Peter Burke, (ed.), New Perspectives on Historical Writing (Uni-versity Park, Pennsylvania University Press, 1992) and History andSocial Theory (Ithaca, Cornell University Press, 1993). Among the clas-sic expositors of social theory history is Cli!ord Geertz, see his ‘ThickDescription: Toward an Interpretive Theory of Culture’, and ‘DeepPlay: Notes on the Balinese Cockfight’, in The Interpretation of Cul-tures (New York, Basic Books, 1973), pp. 3–31, 412–454 and LocalKnowledge: Further Essays in Interpretative Anthropology (New York,Basic Books, 1983). Other examples of this mainstream are FernandBraudel, On History (London, Weidenfeld and Nicolson, 1980). Morerecent examples are Laura Lee Downs, Writing Gender History(London, Arnold, 2004), Nancy Partner (ed.), Writing Medieval His-tory (London, Arnold, 2005) and Garthine Walker (ed.), Writing EarlyModern History (London, Arnold, 2005).

On narrative and the character of writing history it is advisable tostart with W.B. Gallie, Philosophy and the Historical Understanding(New York, Schocken Books, second edition, 1968) and Peter Gay,Style in History: Gibbon, Ranke, Macaulay, Burckhardt (New York,Basic Books, 1974). Also helpful is R. Canary and H. Kozicki (eds), TheWriting of History: Literary Form and Historical Understanding (Madi-son, University of Wisconsin Press, 1978), Arthur Danto, Narrationand Knowledge (New York, Columbia University Press, 1985), DavidCarr, Time, Narrative and History (Bloomington, Indiana UniversityPress, 1986), M.C. Lemon, The Discipline of History and the History ofThought (London, Routledge, 1995) and F.R. Ankersmit and HansKellner (eds) A New Philosophy of History (Chicago, University ofChicago Press, 1995). There is no doubt that the key theorist on narra-tive, representation and history working today is Frank R. Ankersmit.See, for example, Sublime Historical Experience (Stanford, Stanford

Guide to further reading 211

University Press, 2005), ‘Invitation to Historians’, Rethinking History:The Journal of Theory and Practice, Vol. 7, No. 3, 2003, pp. 413–439,Historical Representation (Stanford, Stanford University Press, 2001)and History and Tropology: The Rise and Fall of Metaphor (Berkeley,University of California Press, 1994).

Historians are only slowly becoming familiar with the nature of nar-rative as found in literature. Among the key theorists are Gérard Gen-ette and Seymour Chatman. Gérard Genette maintains that all modesof representation include fictional narrative devices. See his NarrativeDiscourse, trans. Jane E. Lewin (Oxford, Basil Blackwell, [1972] 1986)and Narrative Discourse Revisited, trans. Jane E. Lewin (Ithaca, CornellUniversity Press, [1983] 1990). Seymour Chatman has much that isuseful to say about the nature of narrative, specifically the distinctionbetween story and discourse. See his Story and Discourse: NarrativeStructure in Fiction and Film (Ithaca and London, Cornell UniversityPress, 1978). The other important analysis of the universal functioningof narrative is that provided by Jerome Bruner in his Acts of Meaning(Cambridge, Massachusetts, Harvard University Press, 1990).

There are several important early articles in this area; see, forexample, Lawrence Stone, ‘The Revival of Narrative’, Past & Present,No. 85, 1979, pp. 3–24. For a Marxist response, see E. Hobsbawm,‘Some Comments’, Past and Present, No. 86, 1980, pp. 3–8 and DavidCarr, ‘Narrative and the Real World: An Argument for Continuity’,History and Theory, Vol. 25, No. 2, 1986, pp. 117–131. One of the mostinfluential pieces is John E. Toews, ‘Intellectual History after the Lin-guistic Turn: The Autonomy of Meaning and the Irreducibility ofExperience’, American Historical Review, Vol. 92, No. 4, October, 1987,pp. 879–907. Further important comments on history after the lin-guistic turn can be found in the American Historical Review Forum‘Intellectual History and the Return of Literature’, American HistoricalReview, Vol. 94, No. 3, June, 1989, pp. 581–698. To fully understand thecontribution of Lawrence Stone to the issue of postmodernism andhistory see his ‘History and Post-Modernism’, Past & Present, No. 131,1991, pp. 217–218 and ‘History and Post-Modernism’, Past & Present,No. 135, 1992, pp. 187–194. Also of value is Perez Zagorin, ‘Histori-ography and Postmodernism: Reconsiderations’, History and Theory,Vol. 29, No. 3, 1990, pp. 263–274, Andrew P. Norman, ‘Telling It Like ItWas: Historical Narratives on Their Own Terms’, History and Theory,Vol. 30, No. 2, 1991, pp. 119–135 and Gabrielle M. Spiegel ‘History andPost-Modernism,’ Past and Present, No. 135, 1992, pp. 197–198. Morerecent is Dominick LaCapra, ‘History, Language and Reading: Waitingfor Crillon’, American Historical Review, Vol. 100, No. 3, June, 1995,

212 Guide to further reading

pp. 799–828, and in the same issue Dorothy Ross’ examination of theemplotting of American history, ‘Grand Narrative in American Histor-ical Writing: From Romance to Uncertainty’, pp. 651–677. The mostrecent useful journal articles include Oliver Daddow ‘No PhilosophyPlease, We’re Historians’, Rethinking History: The Journal of Theoryand Practice, Vol. 9, No. 1, 2005, pp. 105–109, Haikki Saari ‘On FrankAnkersmit’s Postmodernist Theory of Historical Narrativity’, Rethink-ing History: The Journal of Theory and Practice, Vol. 9, No. 1, 2005, pp.5–21, and the reply by Frank R. Ankersmit, ‘Reply to Professor Saari’,Rethinking History: The Journal of Theory and Practice, Vol. 9, No. 1,2005, pp. 23–33. On ethics and the historian, see the themed issueon ‘Historians and Ethics’, History and Theory, Vol. 43, No. 4, 2004,pp. 1–164.

This book has argued that central to the relationship between post-modernist intellectual developments in representing the past and writ-ing history is the work of Michel Foucault and Hayden White. MichelFoucault’s key texts include ‘The Order of Discourse’, Inaugural Lec-ture at the College de France, 2 December 1970, The Archaeology ofKnowledge (New York, Harper and Row, 1972), The Order of Things:An Archaeology of the Human Sciences (New York, Random House,1973), Madness and Civilization: A History of Insanity in the Age ofReason (London, Tavistock, 1973), The Birth of the Clinic (New York,Vintage Books, 1975), Counter-Memory, Practice: Selected Essays andInterviews (Ithaca, Cornell University Press, 1979), Power/Knowledge:Selected Interviews and Other Writings (Brighton, Harvester Press,1980). The best commentaries on Foucault are J.G. Merquior, Foucault(London, Fontana, 1985), Jan Goldstein, Foucault and the Writing ofHistory (Oxford, Basil Blackwell, 1994) and Mitchell Dean, Critical andE!ective Histories: Foucault’s Methods and Historical Sociology (Lon-don, Routledge, 1994). An excellent introduction to Foucault as anhistorian is provided by Hayden White, ‘Structuralism and PopularCulture’, Journal of Popular Culture, Vol. 7, 1974, pp. 759–775 and‘Foucault Decoded: Notes From Underground’, History and Theory,Vol. 12, 1973, pp. 23–54. A very comprehensive list of Foucault’s worksare provided in James Bernauer and Thomas Keenan, ‘The Works ofMichel Foucault, 1954–1984’, in James Bernauer and David Rasmussen(eds), The Final Foucault (Cambridge, Massachusetts, MIT Press,1988). Also of value in exploring Foucault are Hubert L. Dreyfus andPaul Rabinow, Michel Foucault: Beyond Structuralism and Hermeneu-tics (Brighton, Harvester Press, second edition, 1983); Mark Poster,Foucault, Marxism and History (London, Polity Press, 1984), and his‘The Reception of Foucault by Historians’, Journal of the History of

Guide to further reading 213

Ideas, Vol. 48, 1987, pp. 117–141, Gary Gutting, The Cambridge Com-panion to Foucault (Cambridge, Cambridge University Press, 1994) andAlan Sheridan, Michel Foucault, The Will to Truth (London, Routledge,reprint, 1994). In another American Historical Review Forum onRussian history, Foucault’s historical method came under closescrutiny; see the articles by Laura Engelstein, Rudy Koshar and JanGoldstein, American Historical Review Forum, Vol. 98, No. 2, April,1993, pp. 338–381. See also Gerard Noiriel’s examination of the contri-bution of Foucault to historical understanding in his ‘Foucault andHistory: The Lessons of a Disillusion’, Journal of Modern History, Vol.66, September, 1994, pp. 547–568.

Hayden White’s contribution to the study of the past is to be found inhis four key texts Metahistory: The Historical Imagination in NineteenthCentury Europe (Baltimore, Johns Hopkins University Press, 1973),Tropics of Discourse: Essays in Cultural Criticism (Baltimore, JohnsHopkins University Press, 1978), The Content of the Form: NarrativeDiscourse and Historical Representation (Baltimore, Johns HopkinsUniversity Press, 1987) and Figural Realism: Studies in the MimesisE!ect (Baltimore, Johns Hopkins University Press, 1998). For examplesof the debate on the Whitean approach to history as a way of knowing,see Arthur Marwick, ‘Two Approaches to Historical Study: The Meta-physical (Including Postmodernism) and the Historical’, Journal ofContemporary History, Vol. 30, No. 1, January, 1995, pp. 5–36, andWhite’s riposte ‘Response to Arthur Marwick’, Journal of Contempor-ary History, Vol. 30, No. 2, April, 1995, pp. 233–246. See also HaydenWhite, ‘An Old Question Raised Again: Is Historiography Art orScience? (Response to Iggers)’, Rethinking History: The Journal ofTheory and Practice, Vol. 4, No. 3, 2000, pp. 391–406. The engagementwith White can also be seen in Hans Kellner, ‘White’s LinguisticHumanism’, History and Theory, Beiheft 19, 1980, pp. 1–29, GregorMcLennan, ‘History and Theory: Contemporary Debates andDirections’, Literature and History, Vol. 10, No. 2, Autumn, 1984, pp.139–164, Paul A. Roth, ‘Hayden White and the Aesthetics of Histori-ography’, History of the Human Sciences, Vol. 5, 1992, pp. 17–35 andThe Routledge Companion to Historical Studies (London and NewYork, Routledge, revised edition, 2006), and Wulf Kansteiner, ‘HaydenWhite’s Critique of the Writing of History’, History and Theory, Vol.32, No. 3, 1993, pp. 273–295. See also the forthcoming collection onthe future of the past edited by Keith Jenkins, Sue Morgan and AlunMunslow (eds), Manifestos for History (London and New York,Routledge, forthcoming). Keith Jenkins has established a unique repu-tation among history theorists as a critic of epistemological thinking

214 Guide to further reading

and practice. See his ‘Ethical Responsibility and the Historian: On thePossible End of History “of a certain kind” ’, History & Theory, Vol.43, No. 4, 2004, pp. 43–60, Refiguring History: New Thoughts on an OldDiscipline (London and New York, Routledge, 2003), ‘Invitation toHistorians: After History’, Rethinking History: The Journal of Theoryand Practice, Vol. 3, No. 1, 1999, pp. 7–20, Why History? Reflections onthe Possible End of History and Ethics under the Impact of the Post-modern (London and New York, Routledge, 1999), Postmodern HistoryReader (London and New York, Routledge, 1997), On ‘What isHistory?’ (London and New York, Routledge, 1995) and, of course, hisseminal Rethinking History (London and New York, Routledge, [1991]2003).

Guide to further reading 215

Notes

1 INTRODUCTION

1 Quoted in Richard T. Vann, ‘Louis Mink’s Linguistic Turn’, History andTheory, Vol. 26, No. 1, 1987, pp. 1–14. See also Louis Mink, ‘History andFiction as Modes of Comprehension’, New Literary History, Vol. 1, 1970,pp. 541–558. While strongly objecting to Hayden White’s placing of literaryform before historical content as the central organisational feature of writ-ten history, a helpful introduction to the relationship of form and contentin historical explanation is to be found in Saul Friedlander (ed.), Probingthe Limits of Representation: Nazism and the ‘Final Solution’ (Cambridge,Massachusetts, Harvard University Press, 1992). See also Nancy Partner,‘Hayden White: the form of the content’, History and Theory, Vol. 37, No.2, 1998, pp. 162–172; Alun Munslow, The New History (Harlow, Pearson,2003), pp. 112–113.

2 A lucid though unsympathetic introduction to this issue is to be found inAlex Callinicos, Theories and Narratives: Reflections on the Philosophyof History (Cambridge, Polity Press, 1995), Introduction, pp. 2–4. Seealso Keith Jenkins, Re-Thinking History (London, Routledge, 2003 [1991]),pp. 12–15.

3 M.C. Lemon, The Discipline of History and the History of Thought(London, Routledge, 1995), p. 131.

4 Ibid., p. 144.5 Ibid. The debate on history and literary fiction has been explored in depth

in a themed double issue of the journal Rethinking History: The Journal ofTheory and Practice, Vol. 9, No. 2/3, 2005, pp. 141–383.

6 This is a well-established position. See Richard Rorty, Philosophy and theMirror of Nature (Princeton, Princeton University Press, 1979); PeterCharles Ho!er and William W. Stueck, Reading and Writing American His-tory: An Introduction to the Historian’s Craft (Lexington, D.C., Heath,1994); Keith Jenkins, On ‘What is History? (London, Routledge, 1995) andAlun Munslow, The Routledge Companion to Historical Studies (Londonand New York, Routledge, second edition, 2006).

7 Arthur Danto, Narration and Knowledge (New York, Columbia UniversityPress, 1985), p. 202.

8 Lemon, op. cit., p. 133. See also Philip Stewart, ‘This is Not a Book Review:

On Historical Uses of Literature’, Journal of Modern History, Vol. 66, No.3, September 1994, pp. 521–538.

9 This description is to be found in Thomas A. Bailey and David M.Kennedy, The American Pageant (Lexington, D.C., Heath, tenth edition,1994), p. 225.

10 William H. Walsh, ‘Colligatory Concepts in History’, in Patrick Gardiner(ed.), The Philosophy of History (New York, Oxford University Press,1974), p. 136; William Dray, Philosophy of History (Englewood Cli!s,Prentice-Hall, second edition, 1993), pp. 89–113; Hayden White, Metahis-tory: The Historical Imagination in Nineteenth Century Europe (Baltimore,Johns Hopkins University Press, 1973), pp. ix–x.

11 R.G. Collingwood, The Idea of History (originally published 1946, Oxford,Oxford University Press, revised edition 1994), pp. 302, 390–395.

12 Carla Hesse, ‘The New Empiricism’, Cultural and Social History, Vol. 1,2004, pp. 201–207.

13 Ibid.14 Howard Marchitello, What Happens to History: The Renewal of Ethics in

Contemporary Thought (London and New York, Routledge, 2001); FrankR. Ankersmit, ‘In Praise of Subjectivity’, in David Carr, Thomas R. Flynnand Rudolf A. Makkreel (eds), The Ethics of History (Evanston,Northwestern University Press, 2004), pp. 3–27.

15 Neville Kirk, ‘The Continuing Relevance and Engagement of Class’,Labour History Review, Vol. 60, No. 3, winter 1995, pp. 2–15.

16 The term used by the philosopher of history Michael E. Hobart to describethis attention to the role of narrative in writing history is rhetorical con-structionism, while White describes it variously as the ‘metahistorical’ or an‘essentially poetic act’ in which the historian ‘prefigures the historical field’.See Hobart, ‘The Paradox of Historical Constructionism’, History andTheory, Vol. 8, No. 1, 1989, pp. 43–58. The only full application and cri-tique of White’s methodology of history is to be found in Alun Munslow,Discourse and Culture: The Creation of America, 1870–1920 (London,Routledge, 1992). A useful assessment of the role of narrative in writing thepast and other issues concerning the postmodern condition of history is tobe found in Robert F. Berkhofer, Beyond the Great Story: History as Textand Discourse (Cambridge, Massachusetts, Harvard University Press,1995). See also David R. Roberts, Nothing But History: Reconstruction andExtremity After Metaphysics (Berkeley, University of California Press,1995). Michael S. Roth, The Ironist’s Cage: Memory, Trauma, and the Con-struction of History (New York, Columbia University Press, 1995); JoyceAppleby (ed.), Knowledge and Postmodernism in Historical Perspective(London, Routledge, 1996); Keith Jenkins, The Postmodern History Reader(London, Routledge, 1997) and Roger Chartier, On the Edge of the Cli!:History, Language and Practice (Baltimore and London, Johns HopkinsUniversity Press, 1997).

17 W.B. Gallie, Philosophy and the Historical Understanding (New York,Schocken Books, second edition, 1968), p. 105. See also Louis Mink, ‘Nar-rative Form as a Cognitive Instrument’, in R. Canary and H. Kozicki (eds),The Writing of History: Literary Form and Historical Understanding(Madison, University of Wisconsin Press, 1978), pp. 129–149.

18 David Carr, ‘Narrative and the Real World: An Argument for Continuity’,

Notes 217

History and Theory, Vol. 25, No. 2, 1986, pp. 117–131; Michel de Certeau,The Writing of History (trans. Tom Conley, New York, Columbia Uni-versity Press, 1988); Paul Ricoeur, Time and Narrative (Chicago, Universityof Chicago Press, 3 vols, 1984, 1985).

19 Paul Veyne, Writing History: Essays on Epistemology (Middletown, Wes-leyan University Press, 1984); Hayden White, ‘The Historical Text asLiterary Artifact’, in Tropics of Discourse: Essays in Cultural Criticism(Baltimore, Johns Hopkins University Press, 1978), p. 82.

20 Hayden White, The Content of the Form: Narrative Discourse and HistoricalRepresentation (Baltimore, Johns Hopkins University Press, 1987), p. 81.See also the collection by F.R. Ankersmit, History and Tropology: The Riseand Fall of Metaphor (Berkeley, University of California Press, 1994), pp.25–28, and his two articles ‘The Dilemma of Contemporary Anglo-SaxonPhilosophy of History’, pp. 44–74, and ‘Historical Representation’, pp. 97–124, both of which originally appeared in the American philosophy ofhistory journal History and Theory.

21 Michel Foucault, Power/Knowledge (Brighton, Harvester Press, 1981),pp. 131–132.

22 White, Content of the Form, op. cit., p. 87.23 George A. Reisch, ‘Chaos, History, and Narrative’, History and Theory,

Vol. 30, No. 1, 1991, pp. 1–20.24 Peter Novick, That Noble Dream: The ‘Objectivity Question’ and the Ameri-

can Historical Profession (Cambridge, Cambridge University Press, 1988),p. 523.

25 Jean-François Lyotard, The Postmodern Condition (Manchester, Manches-ter University Press, 1984), p. 21.

26 This is a point made by the literary critic Robert Young in his study of thedeconstruction of the concept of ‘the West’ in his book White Mythologies:Writing History and the West (London, Routledge, 1990), pp. 1–20.

27 Jenkins, On ‘What is History?’, op. cit., p. 6.28 F.R. Ankersmit, ‘Historiography and Postmodernism’, History and Theory,

Vol. 28, No. 2, 1989, pp. 137–153.29 Ignacio Olábarri, ‘ “New” New History: A Langue Durée Structure’,

History and Theory, Vol. 34, No. 1, 1995, pp. 1–29.

2 THE PAST IN A CHANGING PRESENT

1 Philosopher of history Christopher Lloyd maintains that ‘The writing ofeconomic and social history is now a multifarious, voluminous, and cac-ophonous business’; see Christopher Lloyd, The Structures of History(Oxford, Basil Blackwell, 1993), p. 66. See also Lynn Hunt, The New Cul-tural History (Berkeley, University of California Press, 1989), Introduction,p. 1; Robert Darnton, ‘Intellectual and Cultural History’, in MichaelKammen (ed.), The Past Before Us: Contemporary Historical Writing in theUnited States (Ithaca, Cornell University Press, 1980), pp. 327–354. See alsoPeter Burke (ed.), New Perspectives on Historical Writing (University Park,Pennsylvania University Press, 1991), p. 1 and History and Social Theory(Ithaca, Cornell University Press, 1992). A basic survey of the varieties ofhistory is to be found in J. Gardiner (ed.), What is History Today? (London,

218 Notes

Humanities Press International, 1988) and more recently Michael Bentley(ed.), Companion to Historiography (New York and London: Routledge,1997); Keith Jenkins, Why History? Reflections on the Possible End of His-tory and Ethics under the Impact of the Postmodern (London and New York,Routledge, 1999); A. Green, and K. Troup (eds), The Houses of History: ACritical Reader in Twentieth-century History and Theory (Manchester,Manchester University Press, 1999); Ludmilla Jordanova, History in Prac-tice (London, Arnold, 2002); David Cannadine (ed.) What is History Now?(Houndmills, Palgrave Macmillan, 2002); Kevin Passmore ‘Poststructural-ism and History’, in Stefan Berger, Heiko Feldner and Kevin Passmore(eds), Writing History: Theory and Practice (London, Hodder Arnold,2003), pp. 118–140; Keith Jenkins, Refiguring History: New Thoughts on anOld Discipline (London and New York, Routledge, 2003); Donald M.MacRaild and Avram Taylor, Social Theory and Social History (Hound-mills, Palgrave Macmillan, 2004); Alun Munslow and Robert A. Rosen-stone (eds), Experiments in Rethinking History (London and New York,Routledge); Peter Lambert and Phillipp Schofield (eds), Making History:An Introduction to the History and Practices of a Discipline (London andNew York, Routledge, 2004); Keith Jenkins and Alun Munslow (eds),The Nature of History Reader (London and New York, Routledge, 2004);Willie Thompson, Postmodernism and History (Houndmills, PalgraveMacmillan, 2004); Keith Jenkins, ‘Ethical Responsibility and The His-torian: On the Possible End of History “of a certain kind” ’, History &Theory, Vol. 43, No. 4, 2004, pp. 43–60; Callum G. Brown, Postmodern-ism for Historians (Harlow, Pearson Longman, 2005); Gabrielle Spiegel,Practicing History (New York and London, Routledge, 2005); BeverleySouthgate, What is History For? (New York and London, Routledge,2005); Alun Munslow, Narrative and History (Houndmills, PalgraveMacmillan, forthcoming); Keith Jenkins, Sue Morgan and Alun Munslow(eds), Manifestos for History (London and New York, Routledge,forthcoming).

2 This debate between postmodernity and history is now well established. SeeFrank R. Ankersmit, ‘The Reality E!ect in the Writing of History: TheDynamics of Historical Topology’, in History and Tropology: The Rise andFall of Metaphor (Berkeley, University of California Press, 1994), pp. 125–161; Gertrude Himmelfarb, ‘Some Reflections on the New History’, Ameri-can Historical Review, Vol. 94, No. 3, June 1989, pp. 661–670; LawrenceStone, ‘History and Post-Modernism’, Past and Present, No. 131, May1991, pp. 217–218; C. Behan McCullagh, ‘Metaphor and Truth in History’,Clio, Vol. 23, No. 1, Fall 1993, pp. 23–49; Elizabeth Tonkin, ‘History andthe Myth of Realism’, in Raphael Samuel and Paul Thompson (eds), TheMyths We Live By (London, Routledge, 1990), pp. 25–35; Philippe Carrard,Poetics of the New History: French Historical Discourse from Braudel toChartier (Baltimore, Johns Hopkins University Press, 1992); Alun Muns-low, Discourse and Culture: The Creation of America, 1870–1920 (London,Routledge, 1992); Barbara Melosh (ed.), Gender and American HistorySince 1890 (London, Routledge, 1993); Alex Callinicos, Theories and Narra-tives: Reflections on the Philosophy of History (Oxford, Oxford UniversityPress, 1995) and Keith Jenkins, On ‘What is History?’ (London,Routledge, 1995).

Notes 219

3 Peter Gay, Style in History: Gibbon, Ranke, Macaulay, Burckhardt (NewYork, Basic Books, 1974), p. 3.

4 G.R. Elton, The Practice of History (New York, Crowell, 1967); John Tosh,The Pursuit of History (London, Longman, second edition, 1991); J.H.Hexter, Re-Appraisals in History (Evanston, Northwestern UniversityPress, 1961).

5 Marshal Sahlins, Historical Metaphors and Mythical Realities (Ann Arbor,University of Michigan Press, 1981), Islands of History (Chicago, Uni-versity of Chicago Press, 1985), Boundaries: The Making of France andSpain in the Pyranees (Berkeley, University of California Press, 1989);Anthony Giddens, New Rules of Sociological Method: A Positive Critique ofInterpretative Sociologies (New York, Basic Books, 1976); Cli!ord Geertz,‘Thick Description: Toward an Interpretive Theory of Culture’ and ‘DeepPlay: Notes on the Balinese Cockfight’, in The Interpretation of Cultures(New York, Basic Books, 1973), pp. 3–31, 412–454, and Local Knowledge:Further Essays in Interpretative Anthropology (New York, Basic Books,1983).

6 Harvey Kaye, The British Marxist Historians: An Introductory Analysis(New York, Polity Press, 1984) and The Education of Desire: Marxists andthe Writing of History (London, Routledge, 1992).

7 For a basic introduction see Dominick LaCapra and Steven Kaplan (eds)Modern European Intellectual History: Reappraisals and New Perspectives(Ithaca, Cornell University Press, 1982); Dominick LaCapra, RethinkingIntellectual History: Texts, Contexts, Language (Ithaca, Cornell UniversityPress, 1983); David Harlan, ‘Intellectual History and the Return of Litera-ture’, a contribution that lent its title to the AHR Forum, American Histor-ical Review, Vol. 94, No. 3, June 1989, p. 585; Joan W. Scott. Gender and thePolitics of History (New York, Columbia University Press, 1988) and ‘His-tory in Crisis? The Others’ Side of the Story’, AHR Forum, AmericanHistorical Review, Vol. 94, No. 3, June 1989, pp. 680–692; Stephen Bann,The Clothing of Clio: A Study of the Representation of History in NineteenthCentury Britain and France (Cambridge, Cambridge University Press, 1984)and Roger Chartier, On the Edge of the Cli!: History, Language and Prac-tice (Baltimore and London, Johns Hopkins University Press, 1997). Morerecently, see Frank R. Ankersmit.

8 Tosh, The Pursuit of History, op. cit., p. 48.9 G.R. Elton, Return to Essentials (Cambridge, Cambridge University Press,

1991), pp. 6, 77–98.10 Ibid., p. 12.11 Chris Lorenz, ‘Historical Knowledge and Historical Reality: A Plea for

“Historical Realism” ’, History and Theory, Vol. 33, No. 3, 1994, pp. 297–327.12 Elton, Return to Essentials, op. cit., p. 67.13 Ibid., pp. 67–68.14 Ibid., p. 10.15 Arthur Marwick, The Nature of History (London, Macmillan, third edi-

tion, 1989), pp. 105–106 and also see his much updated version The NewNature of History: Knowledge, Evidence, Language (Houndmills, Palgrave,2001).

16 Lawrence Stone, ‘Dry Heat, Cool Reason: Historians Under Siege in Eng-land and France’, Times Literary Supplement, 31 January 1992.

220 Notes

17 Burke (ed.), New Perspectives, op. cit., pp. 2, 9.18 Mark Cousins, ‘The Practice of Historical Investigation’, in Derek

Attridge, Geo! Bennington and Robert Young (eds), Post-Structuralismand the Question of History (Cambridge, Cambridge University Press,1987), pp. 126–136.

19 Lawrence Stone, ‘The Revival of Narrative’, Past and Present, No. 85, 1979,pp. 3–24. For a Marxist constructionist response see E. Hobsbawm, ‘SomeComments’, Past and Present, No. 86, 1980, pp. 3–8.

20 Lawrence Stone, ‘History and Post-Modernism’, Past and Present, No. 131,1991, pp. 217–218.

21 Lawrence Stone, ‘History and Post-Modernism’, Past and Present, No. 135,1992, pp. 187–194.

22 Roger Chartier, Cultural History: Between Practices and Representations(Cambridge, Cambridge University Press, 1988), p. 42 and On the Edge ofthe Cli!, op. cit., pp. 28–38.

23 A sound introduction to the history and impact of all major aspects ofpostmodernism is Hans Bertens, The Idea of the Postmodern: A History(London, Routledge, 1995), pp. 45, 67, 71–74. See also Alun Munslow, TheNew History (Harlow, Pearson Longman, 2003); Ernst Breisach, On theFuture of History: The Postmodernist Challenge and its Aftermath (Chicagoand London, University of Chicago Press, 2003) for a realist view.

24 Chartier, Cultural History, op. cit., p. 43.25 Jacques Derrida, Of Grammatology (trans. G.C. Spivak, Baltimore, Johns

Hopkins University Press, 1976), Writing and Di!erence (trans. A. Bass,Chicago, University of Chicago Press, 1978), ‘Di!erence’, Speech and Phe-nomena: and Other Essays on Husserl’s Theory of Signs (trans. David B.Allison, Evanston, Northwestern University Press, 1973), pp. 129–160.

26 A useful summary of constructionism is provided by Michael Stanford in ACompanion to History (Oxford, Basil Blackwell, 1994), pp. 128–129. Bar-bara Melosh is very much aware that in her book Gender and AmericanHistory Since 1890 she has edited a collection that is epistemologicallyself-conscious, as she says ‘these essays demonstrate the influence ofpost-structuralist attention to language’, Melosh, op. cit., p. 5.

27 Raymond Williams, Keywords (Oxford, Oxford University Press, 1983),pp. 304–306.

28 Ferdinand de Saussure, Course de Linguistic Générale (1916, trans. WadeBaskin, London, Fontana, 1959). See also Tim Dant, Ideology andDiscourse (London, Routledge, 1991), p. 101.

29 William Pencak, ‘History and Semiotics’, themed issue in The AmericanJournal of Semiotics, Vol. 12, Nos. 1–4, 1995/98.

30 On this important issue see Christopher Norris, Deconstruction: Theory andPractice (London, Methuen, 1982), pp. 1–55. A number of philosophers ofhistory and practising historians have explored the nature of narrative ashistorical explanation; see, for example, William H. Walsh, An Introductionto Philosophy of History (London, Hutchinson, 1958) and Leon Goldstein,Historical Knowing (Austin, University of Texas, 1976). See the excellentsurvey in Geo!rey Roberts, The History and Narrative Reader (London andNew York, Routledge, 2001).

31 Roland Barthes, Mythologies (London, Jonathan Cape, 1972), Elements ofSemiology (New York, Hill & Wang, 1967), S/Z (New York, Hill & Wang,

Notes 221

1975) and Image–Music–Text (New York, Hill & Wang, 1977). This issuewill be taken up further below.

32 Frank R. Ankersmit’s key texts are: ‘Reply to Professor Saari’, RethinkingHistory: The Journal of Theory and Practice, Vol. 9, 2005, pp. 23–33; Sub-lime Historical Experience (Stanford, Stanford University Press, 2005);‘Invitation to Historians’, Rethinking History: The Journal of Theory andPractice, Vol. 7, 2003, pp. 413–439; ‘Pygmalion. Rousseau and Diderot ontheatrical representation’, Rethinking History; The Journal of Theory andPractice, Vol. 7, 2003, pp. 315–341; Political Representation (Stanford, Stan-ford University Press, 2002); Historical Representation (Stanford, StanfordUniversity Press, 2001); ‘Exchanging Ideas’ (with Mark Bevir) in Rethink-ing History: The Journal of Theory and Practice, Vol. 4, 2000, pp. 351–372;‘Hayden White’s appeal to the historians’, History and Theory, Vol. 37,1998, pp. 182–193; ‘Danto on Representation, Identity, and Indiscernibles’,Theme Issue: History and Theory, Vol. 37, 1998, pp. 44–70; History andTropology: The Rise and Fall of Metaphor (Berkeley, University of Califor-nia Press, 1994); ‘Historiography and Postmodernism’, History and Theory,Vol. 28, 1989, pp. 137–153; Narrative Logic: A Semantic Analysis of theHistorian’s Language (The Hague, Martinus Nijho!, 1983); and Frank R.Ankersmit and Hans Kellner (eds), A New Philosophy of History (Chicago,University of Chicago Press, 1995).

33 The term new historicism emerged in Michael McCanles, ‘The AuthenticDiscourse of the Renaissance’, Diacritics, Vol. 10, No. 1, Spring 1980, pp.77–87. The phrase was recoined by Stephen Greenblatt in his essay ‘TheForms of Power and the Power of Forms in the Renaissance’, Genre, Vol.15, Nos 1–2, 1982, pp. 1–4, and has been subsequently elaborated in Green-blatt’s Shakespearean Negotiations: The Circulation of Social Energy inRenaissance England (Berkeley, University of California Press, 1988). In1989 Greenblatt suggested that the movement could be defined as ‘anopenness to the theoretical ferment of the last few years’ and that thisopenness ‘is what distinguishes the new historicism from the positivist his-torical scholarship of the early twentieth century’, Stephen Greenblatt,‘Towards a Poetics of Culture’, in H. Aram Veeser (ed.), The New Histori-cism (London, Routledge, 1989), pp. 1–14. For an alternative definitionthat stresses new historicism as ‘the next step past deconstructionism’, seeJames A. Winn, ‘An Old Historian Looks at the New Historicism’,Comparative Studies in Society and History, Vol. 35, No. 4, October 1993,pp. 859–870.

34 Veeser (ed.), The New Historicism, op. cit., Introduction, passim.35 White, ‘New Historicism: A Comment’, in Veeser (ed.), The New Histori-

cism, op. cit., pp. 293–302.36 Stone, ‘History and Post-Modernism’, Past and Present, No. 131, loc. cit.37 Veeser (ed.), The New Historicism, op. cit., Introduction, p. xi.38 Gay, Style in History, op. cit., p. 3.39 Williams, Keywords, op. cit., p. 306.40 Dant, Ideology and Discourse, op. cit., p. 7; Munslow, Discourse and

Culture, op. cit., pp. 1–3.41 White, ‘The Historical Text as Literary Artifact’, in Tropics of Discourse:

Essays in Cultural Criticism (Baltimore, Johns Hopkins University Press,1978), p. 82.

222 Notes

42 Carrard, Poetics of the New History, op. cit., pp. 18–19. The dissertationshistoriques is the exacting French equivalent of Ph.D. level historical study.

3 HISTORY AS RECONSTRUCTION/CONSTRUCTION

1 Neville Kirk, ‘The Continuing Relevance and Engagement of Class’,Labour History Review, Vol. 60, No. 3, Winter 1995, p. 4.

2 C. Behan McCullagh, Justifying Historical Descriptions (Cambridge, Cam-bridge University Press, 1984), p. 2.

3 Ibid., p. 4. See also his most recent defence of empiricism and truth TheLogic of History: Putting Postmodernism in Perspective (London and NewYork, Routledge, 2004) and his The Truth of History (London andNew York, 1998).

4 C. Behan McCullagh, ‘Can Our Understanding of Old Texts be Objective?’,History and Theory, Vol. 30, No. 3, 1991, pp. 302–323; ‘Bias in HistoricalDescription, Interpretation, and Explanation’, History and Theory, Vol. 39,No. 1, 2000, pp. 39–66.

5 McCullagh, Justifying Historical Descriptions, op. cit., p. 6.6 McCullagh, ‘Can Our Understanding’, op. cit., p. 302.7 James T. Kloppenberg outlined a list similar to this in ‘Objectivity and

Historicism: A Century of American Historical Writing’, AmericanHistorical Review, Vol. 94, No. 4, October 1989, pp. 1011–1030.

8 Joyce Appleby, Lynn Hunt and Margaret Jacob, Telling the Truth AboutHistory (New York, Norton, 1994), p. 248 and Joyce Appleby (ed.), Knowl-edge and Postmodernism in Historical Perspective (London, Routledge,1996), p. 14.

9 Appleby et al., Telling the Truth, op. cit., p. 249.10 Arthur Marwick, The Nature of History (London, Macmillan, third edi-

tion, 1989), p. 21; and also The New Nature of History, op. cit., passim.11 Arthur Marwick, ‘Two Approaches to Historical Study. The Metaphysical

(Including Postmodernism) and the Historical’, Journal of ContemporaryHistory, Vol. 30, No. 1, January 1995, pp. 5–36.

12 Keith Jenkins and Alun Munslow (eds), The Nature of History Reader(London and New York, Routledge, 2004).

13 Edward Royle, Modern Britain: A Social History, 1750–1997 (London,Arnold, [1987] 1998), pp. 120–125.

14 Michael A.R. Graves, Elizabethan Parliaments, 1559–1601 (London,Pearson Education, [1987] 1996).

15 G.R. Elton, Return to Essentials (Cambridge, Cambridge University Press,1991), p. 51.

16 John Tosh, The Pursuit of History (London, Longman, second edition,1991), p. 53.

17 Elton, Return to Essentials, op. cit., p. 52.18 Ibid., p. 55.19 Ibid., p. 62.20 Ibid., p. 66.21 Ibid., p. 70.22 Michael A. Stanford, A Companion to History (Oxford, Basil Blackwell,

1994), p. 124.

Notes 223

23 David Hollinger, ‘The Return of the Prodigal: The Persistence of HistoricalKnowing’, American Historical Review, Vol. 94, No. 3, June 1989, p. 613.

24 Elton, Return to Essentials, op. cit., p. 11.25 E.H. Carr, What Is History? (London, Penguin, second edition, 1987),

p. 65.26 Ibid., p. 22.27 Peter Burke, History and Social Theory (Ithaca, Cornell University Press,

1993), p. 1.28 Ibid., p. 28.29 Ibid., p. 29.30 Appleby et al., Telling the Truth, op. cit., p. 304.31 Alex Callinicos, Theories and Narratives: Reflections on the Philosophy of

History (Cambridge, Polity Press, 1995), p. 77.32 Appleby et al., Telling the Truth, op. cit., p. 304.33 Callinicos, Theories and Narratives, op. cit., p. 82.34 James Harvey Robinson, The New History: Essays Illustrating the Modern

Historical Outlook (New York, Free Press, 1965).35 Frederick Jackson Turner quoted in Peter Novick, That Noble Dream: The

Objectivity Question and the American Historical Profession (Cambridge,Cambridge University Press, 1988), p. 92.

36 Philippe Carrard, Poetics of the New History: French Historical Discoursefrom Braudel to Chartier (Baltimore, Johns Hopkins University Press,1992), pp. 1–28.

37 Christopher Lloyd, The Structures of History (Oxford, Basil Blackwell,1993), p. 83. See also Christopher Lloyd, ‘History and the Social Sciences’,in Berger et al., Writing History, op. cit., pp. 83–103.

38 Carrard, Poetics of the New History, op. cit., p. 31.39 Carl Hempel, ‘The Function of General Laws in History’, The Journal of

Philosophy, Vol. 34, 1942, reprinted in Patrick Gardiner (ed.), Theories ofHistory (New York, Free Press, 1959).

40 Ibid., p. 351. See also Murray G. Murphey, ‘Explanation, Causes, andCovering Laws’, History and Theory, Beiheft 25, 1986, pp. 43–57.

41 Anthony Giddens, Profiles and Critiques in Social Theory (Berkeley, Uni-versity of California Press, 1982) and Social Theory and Modern Sociology(Stanford, Stanford University Press, 1987); Ernest Gellner, Culture, Iden-tity and Politics (Cambridge, Cambridge University Press, 1987); CharlesTilly, From Mobilisation to Revolution (Reading, Massachusetts, Addison-Wesley, 1978) and Big Structures, Large Processes, Huge Comparisons (NewYork, Russell Sage Foundation, 1984); Cli!ord Geertz, The Interpretationof Cultures (New York, Basic Books, 1973) and Local Knowledge (NewYork, Basic Books, 1976); Fernand Braudel, The Mediterranean and theMediterranean World in the Age of Philip II (New York, Harper & Row,1972) and The Identity of France (New York, Harper & Row, 1988–90);Emmanuel Le Roy Ladurie, The Peasants of Languedoc (Paris, Flam-marion, 1969) and Montaillou (New York, G. Braziller, 1978); Robert Darn-ton, The Great Cat Massacre and Other Episodes in French Cultural History(New York, Basic Books, 1985); Roger Chartier, Cultural History: BetweenPractices and Representations (Cambridge, Polity Press, 1988); W.G.Hoskins, The Making of the English Landscape (London, Penguin, 1955);Harry Braverman, Labor and Monopoly Capitalism (New York, Monthly

224 Notes

Review Press, 1974); James Weinstein, The Corporate Ideal in the LiberalState (Boston, Beacon Press, 1968); Gabriel Kolko, The Roots of AmericanForeign Policy (Boston, Beacon Press, 1969); Herbert Gutman, ‘Work,Culture and Society in Industrialising America, 1820–1920’, AmericanHistorical Review, Vol. 78, No. 3, 1973, pp. 531–587; David Montgomery,Workers’ Control in America: Studies in the History of Work, Technologyand Labor Struggles (Cambridge, Cambridge University Press, 1980); EricHobsbawm, The Age of Empire (New York, Pantheon Books, 1987);Eugene Genovese, In Red and Black: Marxian Explorations in Southern andAfro-American History (New York, Vintage Books, 1971); Sheila Row-botham, Hidden From History (London, Pluto Press, 1983) and CatherineHall, White, Male and Middle Class: Explorations in Feminism and History(Cambridge, Polity Press, 1992).

42 J.H. Hexter, ‘The Rhetoric of History’, International Encyclopaedia of theSocial Sciences (1968), first quotation in Novick, That Noble Dream, op.cit., p. 623, and Hexter, The History Primer (New York, Basic Books, 1971),pp. 108, 222.

43 Ibid., pp. 137–138.44 M.C. Lemon, The Discipline of History and the History of Thought

(London, Routledge, 1995), pp. 184–186.45 Hayden White, ‘The Question of Narrative in Contemporary Historical

Theory’, in The Content of the Form: Narrative Discourse and HistoricalRepresentation (Baltimore, Johns Hopkins University Press, 1987), pp. 26–57; Andrew Norman, ‘Telling It Like It Was: Historical Narratives on TheirOwn Terms’, History and Theory, Vol. 30, 1991, pp. 119–135 and WilliamH. Dray, Philosophy of History (Englewood Cli!s, Prentice-Hall, secondedition, 1993), pp. 91–95.

46 Lawrence Stone, ‘Revival of Narrative’, Past and Present, No. 85, 1979, pp.3–4, 19.

47 Ibid., p. 19.48 W.B. Gallie, Philosophy and the Historical Understanding (New York,

Schocken Books, second edition, 1968), p. 105. See also Arthur Danto,Narration and Knowledge (New York, Columbia University Press, 1985).

49 Carrard, Poetics of the New History, op. cit., p. 75.50 Frank R. Ankersmit, ‘Reply to Professor Saari’, Rethinking History: The

Journal of Theory and Practice, Vol. 9, 2005, pp. 23–33; Sublime HistoricalExperience (Stanford, Stanford University Press, 2005); ‘Invitation to His-torians’, Rethinking History: The Journal of Theory and Practice, Vol. 7,2003, pp. 413–439; ‘Pygmalion. Rousseau and Diderot on theatrical repre-sentation’, Rethinking History; The Journal of Theory and Practice, Vol. 7,2003, pp. 315–341; Historical Representation (Stanford, Stanford UniversityPress, 2002); ‘Exchanging Ideas’ (with Mark Bevir) in Rethinking History:The Journal of Theory and Practice, Vol. 4, 2000, pp. 351–372; ‘HaydenWhite’s appeal to the historians’ History and Theory. Vol. 37, 1998, pp.182–193.

51 Appleby et al., Telling the Truth, op. cit., p. 238.52 Ibid., pp. 234–235.53 Stanford, A Companion to History, op. cit., p. 95.54 Ibid., p. 102.55 Ibid., p. 104.

Notes 225

56 Phyllis Deane, The First Industrial Revolution (Cambridge, Cambridge Uni-versity Press, 1965); Clive Trebilcock, The Industrialisation of the Contin-ental Powers 1780–1914 (London, Longman, 1981) and Vicki L. Ruiz andEllen Carol DuBois (eds), Unequal Sisters (London, Routledge, thirdedition, 2000).

57 Elton, Return to Essentials, op. cit., p. 12.58 McCullagh, Justifying Historical Descriptions, op. cit., pp. ix–x.

4 HISTORY AS DECONSTRUCTION

1 Mark Poster, ‘Foucault and History’, Social Research, Vol. 49, 1982, p. 120;Jan Goldstein, Foucault and the Writing of History (Oxford, Basil Black-well, 1994) and Mitchell Dean, Critical and E!ective Histories: Foucault’sMethods and Historical Sociology (London, Routledge, 1994).

2 Allan Megill, ‘Foucault, Structuralism, and the Ends of History’, Journalof Modern History, Vol. 51, September 1979, p. 451.

3 Charles Beard, ‘Written History as an Act of Faith’, American HistoricalReview, Vol. 39, No. 2, 1934, pp. 219–231 and ‘That Noble Dream’,American Historical Review, Vol. 41, No. 1, 1935, pp. 74–87.

4 Rudy Koshar, ‘Foucault and Social History: Comments on “CombinedUnderdevelopment” ’, American Historical Review, Vol. 98, No. 2, April1993, pp. 354–363.

5 Roland Barthes, ‘Le discours de l’histoire’, Information sur les sciencessociales, Vol. 6, No. 4, 1967, pp. 65–75, translated as ‘Discourse of History’with an introduction by Stephen Bann, Comparative Criticism – AYearbook, Vol. 3 (University Park, Pennsylvania University Press, 1981),pp. 3–20.

6 Quoted by Bann in ibid., p. 3.7 Ibid., p. 5.8 Barthes, ‘Discourse of History’, op. cit., p. 7.9 Ibid., p. 11.

10 Barthes, ‘Discourse of History’, op. cit., p. 16.11 Ibid., p. 17. See also Stephen Bann, ‘Analysing the Discourse of History’,

Renaissance and Modern Studies, Vol. 27, 1983, pp. 61–84.12 Barthes, ‘Discourse of History’, op. cit., p. 18. See Richard J. Ellis and Alun

Munslow, ‘Narrative, Myth and the Turner Thesis’, Journal of AmericanCulture, Vol. 9, No. 2, 1987, pp. 9–17 and Alun Munslow, Discourse andCulture: The Creation of America, 1870–1920 (London, Routledge, 1992),pp. 68–88.

13 See Hayden White, ‘The Question of Narrative in Contemporary HistoricalTheory’, History and Theory, Vol. 23, No. 1, 1984, pp. 1–33.

14 Andrew P. Norman, ‘Telling It Like It Was: Historical Narratives on TheirOwn Terms’, History and Theory, Vol. 30, No. 2, 1991, pp. 119–135.

15 Roland Barthes, ‘The Death of the Author’, quoted in David Harlan’s‘Intellectual History and the Return of Literature’, a contribution that lentits title to the AHR Forum, American Historical Review, June 1989, p. 585.

16 F.R. Ankersmit, ‘Historiography and Postmodernism’, History and Theory,Vol. 28, No. 2, 1989, p. 146.

17 Hayden White, ‘The Context in the Text: Method and Ideology in

226 Notes

Intellectual History’, in The Content of the Form: Narrative Discourse andHistorical Representation (Baltimore, Johns Hopkins University Press,1987), p. 192.

18 G.R. Elton, Return to Essentials (Cambridge, Cambridge University Press,1991), p. 49.

19 Hayden White, ‘The Burden of History’, in Tropics of Discourse: Essays inCultural Criticism (Baltimore, Johns Hopkins University Press, 1978), p. 47.

20 R.G. Collingwood, The Idea of History (Oxford, Oxford University Press,revised edition, 1994), pp. 282–302.

21 Ibid., p. 302.22 Ibid.; see Elton’s commentary, Return to Essentials, op. cit., p. 43.23 The original examination of the character of general or covering laws in

historical explanation is to be found in C.G. Hempel, ‘The Function ofGeneral Laws in History’, Journal of Philosophy, Vol. 39, 1942, reprinted inPatrick Gardiner, Theories of History (New York, Free Press, 1959), pp.344–356.

24 Frederick Jackson Turner, Rise of the New West, 1819–1829 (1906), a vol-ume in the series The American Nation; The United States, 1830–1850: TheNation and Its Sections (New York, H. Holt & Co., 1935) with an introduc-tion by Avery Craven; The Frontier in American History (1920, New York,reprinted by Holt, Rinehart & Winston, 1962); Martin Ridge, ‘FrederickJackson Turner, Ray Allen Billington, and Frontier History’, Western His-torical Quarterly, Vol. 19, January 1988, pp. 5–20; Munslow, Discourse andCulture, op. cit., pp. 68–88; John Mack Faragher, ‘The Frontier Trail:Rethinking Turner and Reimagining the American West’, American Histor-ical Review, Vol. 98, No. 1, February 1993, pp. 106–117 and Peter Stoneley,‘Signifying Frontiers’, Borderlines, Vol. 1, No. 3, March 1994, pp. 237–253.

25 Turner, ‘The Significance of the Frontier in American History’, in TheFrontier in American History, op. cit., pp. 2–3.

26 Benedetto Croce, Aesthetics as Science of Expression and General Lin-guistic, translated by Douglas Ainslie with a new Introduction by JohnMcCormick (New Brunswick, Transaction Publishers, 1995).

27 Carl Becker quoted in Peter Novick, That Noble Dream: The ObjectivityQuestion and the American Historical Profession (Cambridge, CambridgeUniversity Press, 1988), p. 98.

28 Karl Popper, The Logic of Scientific Discovery (London, Hutchinson,1959). According to Allan Megill, ‘Recounting the Past: “Description”,Explanation, and Narrative in Historiography’, American HistoricalReview, Vol. 94, No. 3, June 1989, pp. 627–653, ‘the positivist programmestill retains an aura of prestige’ in historical explanation, p. 636. See also‘ “Grand Narrative” and the Discipline of History’, in Frank Ankersmitand Hans Kellner (eds), A New Philosophy of History (Chicago, ChicagoUniversity Press, 1995).

29 Collingwood, The Idea of History, op. cit., p. 130.30 Dorothy Ross, ‘Grand Narratives in American Historical Writing: From

Romance to Uncertainty’, American Historical Review, Vol. 100, No. 3,June 1995, pp. 651–677.

31 Christopher Tilley (ed.), Reading Material Culture (Oxford, BasilBlackwell, 1990), pp. 281–347.

32 Quoted in Norman, ‘Telling It Like It Was’, op. cit., p. 130.

Notes 227

33 White, ‘The Question of Narrative’, op. cit., p. 19.34 Amy J. Elias, ‘Metahistorical Romance, the Historical Sublime, and

Dialogic History’, Rethinking History: The Journal of Theory and Practice,Vol. 9, No. 2/3 2005, pp. 159–172.

35 W.H. Dray, ‘On the Nature and Role of Narrative in Historiography’,History and Theory, Vol. 10, 1970, pp. 153–171.

36 Quoted in Norman, ‘Telling It Like It Was’, op. cit., p. 117.37 Harlan, ‘Intellectual History’, op. cit., p. 600.38 A.R. Louch, ‘History as Narrative’, History and Theory, Vol. 8, 1969,

pp. 54–70.39 William Dray, ‘Mandelbaum on Historical Narrative’, History and Theory,

Vol. 8, 1969, p. 290, quoted in Leon Goldstein, Historical Knowing (Austin,Texas, 1976), p. 140.

40 Ibid., Introduction, p. xix. See also Goldstein, ‘Impediments to Epistem-ology in the Philosophy of History’, History and Theory, Beiheft 25, 1986,pp. 82–100.

41 Ibid., Introduction, pp. xx–xxiii.42 William Gallie, Philosophy and the Historical Understanding (New York,

Schocken Books, second edition, 1968), pp. 105–125 and M.C. Lemon, TheDiscipline of History and the History of Thought (London, Routledge,1995), pp. 42–79.

43 Lemon, The Discipline, op. cit., p. 133.44 Paul Ricoeur, Hermeneutics and the Human Sciences, ed. by J.B. Thompson

(Cambridge, Cambridge University Press, 1981), p. 275.45 Hayden White, ‘Response to Arthur Marwick’, Journal of Contemporary

History, Vol. 30, No. 2, April 1995, pp. 233–246.46 Roland Barthes, ‘Introduction to the Structural Analysis of Narrative’,

quoted in White, ‘The Question of Narrative’, op. cit., p. 1. See also PaulRicoeur, ‘Explanation and Understanding: On Some Remarkable Connec-tions Among the Theory of the Text, Theory of Action, and Theory ofHistory’, quoted in ibid., p. 26. See also Michel Foucault, ‘The Order ofDiscourse’, Inaugural Lecture at the Collège de France, 2 December 1970,The Archaeology of Knowledge (New York, Harper & Row, 1972), TheOrder of Things: An Archaeology of the Human Sciences (New York, Ran-dom House, 1973), Madness and Civilisation: A History of Insanity in theAge of Reason (London, Tavistock, 1973), The Birth of the Clinic (NewYork, Vintage Books, 1975), Language, Counter-Memory, Practice: SelectedEssays and Interviews (Ithaca, Cornell University Press, 1979) and Power/Knowledge: Selected Interviews and Other Writings (Brighton, HarvesterPress, 1980).

47 F.R. Ankersmit, History and Tropology: The Rise and Fall of Metaphor(Berkeley, University of California Press, 1994), p. 83.

48 ‘Unconventional History’, History and Theory (themed issue), Vol. 41,2002, pp. 1–144.

49 Kevin Passmore, ‘Poststructuralism and History’ in Berger et al., WritingHistory, pp. 118–140.

50 Alun Munslow and Robert A. Rosenstone (eds), Experiments in RethinkingHistory (London and New York, Routledge, 2004).

51 Hayden White, ‘Structuralism and Popular Culture’, Journal of PopularCulture, Vol. 7, 1974, pp. 759–775, ‘The Tropics of History: The Deep

228 Notes

Structure of the New Science’ and ‘Foucault Decoded: Notes From Under-ground’, in Tropics of Discourse, op. cit., pp. 197–217, 230–260; Munslow,Discourse and Culture, op. cit., pp. 1–4.

52 ‘Otherness’ as a historical construct has been much explored by decon-structionist historians and critical theorists like Luce Irigaray, This Sexwhich is not One (Ithaca, Cornell University Press, 1979), and also see Hay-den White, Metahistory: The Historical Imagination in the Nineteenth Cen-tury (Baltimore, Johns Hopkins University Press, 1973), pp. 133–425. Aculture results from the bargaining between dominant and subordinategroups and is represented through the metaphors, icons and imagesemployed by such groups. On tropes and their cultural significance see PaulRicoeur, The Rule of Metaphor: Multi-Disciplinary Studies of the Creationof Meaning in Language (Toronto, Toronto University Press, 1978), pp. 44–64 and Stephen Bann, The Clothing of Clio: A Study of the Representationof History in Nineteenth Century Britain and France (Cambridge, Cam-bridge University Press, 1984).

53 White, The Content of the Form, op. cit., Introduction, pp. 1–23.54 White, Tropics of Discourse, op. cit., Introduction, p. 19.55 Roland Barthes, Mythologies (London, Cape, 1972), p. 129.56 The anthropologist Cli!ord Geertz has been one of the main advocates of

the textual model for understanding culture. See his ‘Thick Description:Toward an Interpretative Theory of Culture’ and his ‘Deep Play: Notes onthe Balinese Cockfight’ in his collection The Interpretation of Cultures (NewYork, Basic Books, 1973), pp. 3–30, 412–453.

57 White, ‘The Context in the Text’, in The Content of the Form, op. cit., p. 188.58 White, ‘The Absurdist Moment in Contemporary Literary Theory’, in

Tropics of Discourse, op. cit., pp. 261–282.59 White, ‘Historicism, History, and the Figurative Imagination’, in ibid.,

p. 117.60 Ibid.61 Hayden White, ‘The Metaphysics of Narrativity: Time and Symbol in

Ricoeur’s Philosophy of History’, in The Content of the Form, op. cit., p. 173.62 Ricoeur, Hermeneutics and the Human Sciences, op. cit., p. 279. See also

Robert Scholes and Robert Kellogg, The Nature of Narrative (New York,Oxford University Press, 1966) and Saul Friedlander (ed.), Probing theLimits of Representation: Nazism and the ‘Final Solution’ (Cambridge,Massachusetts, Harvard University Press, 1992).

63 Hayden White, ‘The Metaphysics of Narrativity’, op. cit., p. 173.64 Ibid., p. 181.65 Ibid.

5 WHAT IS WRONG WITH DECONSTRUCTIONIST HISTORY?

1 Fred A. Olafson, ‘Hermeneutics, “Analytical” and “Dialectical” ’, Historyand Theory, Beiheft 25, 1986, pp. 28–42.

2 John Tosh, The Pursuit of History (London, Longman, second edition,1991), p. 108.

3 Joyce Appleby, Lynn Hunt and Margaret Jacob, Telling the Truth AboutHistory (New York, Norton, 1994), pp. 160–197.

Notes 229

4 T.S. Kuhn, The Structure of Scientific Revolutions (Chicago, University ofChicago Press, 1961).

5 Appleby et al., Telling the Truth, op. cit., pp. 195–196.6 Frank R. Ankersmit and Mark Bevir, ‘Exchanging Ideas’, Rethinking

History: The Journal of Theory and Practice, Vol. 4, 2000, pp. 351–372.7 Michel Foucault, ‘What is Enlightenment?’, in Paul Rabinow, The Foucault

Reader (New York, Random House, 1984), pp. 32–50.8 Appleby et al., Telling the Truth, op. cit., p. 212.9 Linda Gordon, ‘Comments on That Noble Dream’, American Historical

Review, Vol. 96, No. 3, June 1991, pp. 683–687.10 G.R. Elton, Return to Essentials (Cambridge, Cambridge University Press,

1991), p. 29. What is intended to be an accessible guide to key concepts inhistory, like truth, form and content, objectivity, event, knower and known,etc. is provided in Munslow, Routledge Companion, op. cit.

11 Michael Stanford, A Companion to the Study of History (Oxford, BasilBlackwell, 1994), p. 91.

12 Arthur Marwick, ‘Two Approaches to Historical Study: The Metaphysical(Including Postmodernism) and the Historical’, Journal of ContemporaryHistory, Vol. 30, No. 1, January 1995, pp. 18–20.

13 Appleby et al., Telling the Truth, op. cit., p. 227.14 James T. Kloppenberg, ‘Objectivity and Historicism: A Century of Ameri-

can Historical Writing’, American Historical Review, Vol. 94, No. 4, Octo-ber 1989, p. 1017.

15 Richard Rorty, Philosophy and the Mirror of Nature (Princeton, PrincetonUniversity Press, 1980) and Consequences of Pragmatism (Minneapolis,University of Minnesota Press, 1982); Richard J. Bernstein, ‘TheResurgence of Pragmatism’, Social Research, Vol. 59, 1992, pp. 825–826.

16 Kloppenberg, ‘Objectivity and Historicism’, op. cit., p. 1018.17 Quoted in ibid., p. 1020.18 Ellen Nore, ‘Charles A. Beard’s Act of Faith: Context and Content’, The

Journal of American History, Vol. 66, No. 4, March 1980, pp. 850–866 andCharles A. Beard: An Intellectual Biography (Carbondale, Southern IllinoisUniversity Press, 1983).

19 Leon Goldstein, ‘Impediments to Epistemology in the Philosophy of His-tory’, History and Theory, Beiheft 25, 1986, p. 96.

20 Marwick, ‘Two Approaches’, op. cit., pp. 20–23. See also John M. Ellis,Against Deconstruction (Princeton, Princeton University Press, 1989), p. 138.

21 Tosh, The Pursuit of History, op. cit., p. 137.22 Marwick, ‘Two Approaches’, op. cit., p. 21.23 C. Behan McCullagh, ‘Metaphor and Truth in History’, Clio, Vol. 23, No.

1, Fall 1993, p. 36. See also Paul Ricoeur, The Rule of Metaphor: Multi-disciplinary Studies of the Creation of Meaning in Language (London,Routledge, 1994 [1975]).

24 Ibid., p. 37.25 Ibid.26 E.H. Carr, What Is History? (London, Penguin, second edition, 1987),

p. 11.27 Ibid., p. 11.28 R.G. Collingwood, The Idea of History (Oxford, Oxford University Press,

revised edition, 1994), p. 244.

230 Notes

29 Carr, What is History?, op. cit., pp. 12–13.30 See Keith Jenkins’ treatment of the Carr-Elton debate in On ‘What is

History?’ (London, Routledge, 1995), pp. 42–96, passim.31 Peter Gay, Style in History (New York, Norton, 1974), p. 198.32 Ibid., pp. 199, 217; Peter Novick, That Noble Dream: The ‘Objectivity

Question’ and the American Historical Profession (Cambridge, CambridgeUniversity Press, 1988), p. 611.

33 McCullagh, ‘Metaphor and Truth’, op. cit., p. 43.34 Carr, What is History?, op. cit., p. 14.35 Tosh, The Pursuit of History, op. cit., p. 139.36 Elton, Return to Essentials, op. cit., p. 19.37 F.J. Turner, ‘Social Forces in American History’, in The Frontier in Ameri-

can History (New York, Holt, Rinehart & Winston, 1920, reprinted 1962),pp. 311–334. This was the speech he delivered to the American HistoricalAssociation after his election as President of the Association in 1910.

38 Elton, Return to Essentials, op. cit., p. 6.39 Ibid., pp. 9–11.40 Ibid., pp. 15, 19.41 Gertrude Himmelfarb, ‘Some Reflections on the New History’, American

Historical Review, Vol. 94, No. 3, June 1989, p. 665, The New History andthe Old (Cambridge, Massachusetts, Harvard University Press, 1987).

42 Gertrude Himmelfarb, ‘The New History’, New York Times Review ofBooks, Vol. 17, August 1980, p. 3, quoted in Novick, That Noble Dream,op. cit., p. 610.

43 Ibid.44 Lawrence Stone, letter to Harper’s Magazine, Vol. 268, June 1984, pp. 4–5,

quoted in ibid.45 Lawrence Stone, ‘The Revival of Narrative’, Past and Present, No. 85, 1979,

p. 4.46 Ibid., pp. 4–8.47 Ibid., p. 23.48 Ibid., p. 19.49 Lawrence Stone, ‘History and Post-Modernism’, Past and Present, No. 135,

May 1992, p. 217.50 Ibid.51 Ibid., pp. 189–190.52 Ibid., p. 192.53 Ibid., pp. 193–194.54 Gabrielle M. Spiegel, ‘History and Post-Modernism’, Past and Present, No.

135, May 1992, pp. 197–198, and Practicing History, op. cit.55 Ibid., p. 203.56 Tosh, The Pursuit of History, op. cit., p. 138.57 Ibid., p. 139.58 A.J.P. Taylor, ‘Fiction in History’, Times Literary Supplement, 23 March

1973, p. 327.59 Ibid., p. 328.60 Ibid.61 Robert F. Berkhofer, Beyond the Great Story: History as Text and

Discourse (Cambridge, Massachusetts, Harvard University Press, 1995),pp. 38–50.

Notes 231

62 Richard T. Vann, ‘Theory and Practice in Historical Study’, Guide to His-torical Literature, Beth Norton and Pamela Gerardi (eds) (New York,American Historical Association, 1995), pp. 1–4.

63 Ibid., p. 4.64 Collingwood, The Idea of History, op. cit., p. 391.65 C. Behan McCullagh, Justifying Historical Descriptions (Cambridge, Cam-

bridge University Press, 1984), pp. 8–10 and ‘Metaphor and Truth’, op. cit.,pp. 43–44.

66 Olafson, ‘Hermeneutics’, op. cit., p. 40.67 David Carroll, ‘Poetics, Theory, and the Defence of History’, Clio, Vol. 22,

No. 3, 1993, pp. 273–289, a review of Philippe Carrard’s Poetics of the NewHistory, op. cit.

68 Ibid., p. 277.69 Ibid., p. 289. See also William Cronon, ‘A Place for Stories: Nature, His-

tory, and Narrative’, Journal of American History, Vol. 78, March 1992, pp.1347–1376, who very much doubts that radically di!erent multiple inter-pretations using the same evidence are viable.

70 Appleby et al., Telling the Truth, op. cit., pp. 254–257.71 Arthur Danto, Narration and Knowledge (New York, Columbia University

Press, 1985), p. 177.72 Andrew P. Norman, ‘Telling It Like It Was: Historical Narratives On Their

Own Terms’, History and Theory, Vol. 30, No. 2, 1991, pp. 133–134.73 Appleby et al., Telling the Truth, op. cit., p. 229.74 Ibid.75 Ibid., pp. 229–230.76 Ibid., p. 230.77 Eric Hobsbawm in Felix Gilbert and E.R. Graubard (eds), Historical

Studies To-Day (New York, Norton, 1972), p. 9, quoted in Stanford, ACompanion, op. cit., p. 106.

78 Alasdair MacIntyre, ‘Epistemological Crisis, Dramatic Narrative, and thePhilosophy of Science’, The Monist, Vol. 60, 1978, p. 457, quoted inNorman, ‘Telling It Like It Was’, op. cit., p. 131.

79 Alex Callinicos, Theories and Narratives: Reflections on the Philosophy ofHistory (Cambridge, Polity Press, 1995), p. 71. See also Hayden White,‘Historical Emplotment and the Problem of Truth’, in Saul Friedlander(ed.), Probing the Limits of Representation (Cambridge, Massachusetts,Harvard University Press, 1992), pp. 37–53.

80 Collingwood, The Idea of History, op. cit., p. 32.81 James A. Winn, ‘An Old Historian Looks at the New Historicism’,

Comparative Studies in Society and History, Vol. 35, No. 4, October 1993,pp. 867–868.

82 Appleby et al., Telling the Truth, op. cit., p. 251.

6 WHAT IS WRONG WITH RECONSTRUCTIONIST/CONSTRUCTIONIST HISTORY?

1 Jerzy Topolski, ‘Towards an Integrated Model of Historical Explanation’,History and Theory, Vol. 30, No. 3, 1991, pp. 324–338; Brown, Postmodern-ism, op. cit., pp. 26–29, 96–99.

232 Notes

2 Joan W. Scott, ‘History in Crisis? The Others’ Side of the Story’, AmericanHistorical Review, Vol. 94, No. 3, June 1989, pp. 680–692.

3 Hans Kellner, Language and Historical Representation: Getting the StoryCrooked (Madison, University of Wisconsin Press, 1989), p. vii.

4 Gérard Genette, Narrative Discourse, trans. Jane E. Lewin (Oxford, BasilBlackwell, 1986 [1972]) and Narrative Discourse Revisited, trans. Jane E.Lewin (Ithaca, Cornell University Press, 1990 [1983]); Seymour Chatman,Story and Discourse: Narrative Structure in Fiction and Film (Ithaca andLondon, Cornell University Press, 1978); Jerome Bruner in his Acts ofMeaning (Cambridge, Massachusetts, Harvard University Press, 1990);Paul Ricoeur, Time and Narrative, trans. Kathleen McLaughlin and DavidPellauer (Vol. 1) (Chicago, University of Chicago Press, 1984), p. 41.

5 Perez Zagorin, ‘Historiography and Postmodernism: Reconsiderations’,History and Theory, Vol. 29, No. 3, 1990, pp. 263–274.

6 Jörn Rüsen quoting Ranke in ‘Rhetoric and Aesthetics of History: LeopoldVon Ranke’, History and Theory, Vol. 29, No. 2, 1990, pp. 190–204.

7 David A. Hollinger; ‘Postmodernist Theory and Wissenschaftliche Prac-tice’, AHR Forum, American Historical Review, Vol. 96, No. 3, June 1991,pp. 688–692.

8 Mark Bevir, ‘Objectivity in History’, History and Theory, Vol. 33, No. 3,1994, pp. 328–344.

9 Gabrielle M. Spiegel, Romancing the Past: The Rise of Vernacular ProseHistoriography in Thirteenth Century France (Berkeley, University of Cali-fornia Press, 1992) and Carol Douglas Sparks, ‘The Land Incarnate: NavajoWomen and the Dialogue of Colonialism, 1821–1870’, in Nancy Shoe-maker (ed.), Negotiators of Change: Historical Perspectives on Native Amer-ican Women (New York, Routledge, 1995), pp. 135–156.

10 James R. Kincaid, Child Loving: The Erotic Child and Victorian Culture(New York, Routledge, 1992), p. 5.

11 Sparks, ‘The Land Incarnate’, op. cit., pp. 136–137.12 Roger Chartier, Cultural History: Between Practices and Representations

(Cambridge, Polity Press, 1988), p. 42.13 J.G.A. Pocock, Virtue, Commerce and History: Essays on Political Thought

and History Chiefly in the Eighteenth Century (Cambridge, CambridgeUniversity Press, 1985), pp. 8–15.

14 Mark Bevir, ‘The Errors of Linguistic Contextualism’, History and Theory,Vol. 31, No. 3, 1992, pp. 276–298.

15 Peter Burke (ed.), New Perspectives on Historical Writing (Cambridge,Cambridge University Press, 1991), p. 238.

16 Gabrielle M. Spiegel, ‘History and Post-Modernism’, Past and Present,No. 135, May 1992, p. 197.

17 David Harlan, ‘Intellectual History and the Return of Literature’, Ameri-can Historical Review, Vol. 94, No. 3, June 1989, pp. 581–609.

18 Bruno Latour and Steve Woolgar, Laboratory Life: The Social Constructionof Scientific Facts (Los Angeles, Sage, 1979) and also more recently BrunoLatour, Aramis or the Love of Technology, trans. Catherine Porter (Harvard,Harvard University Press, 1996).

19 Harlan, loc. cit., p. 609.20 Ibid. See also David Harlan, The Degradation of American History

(Chicago, University Chicago Press, 1997).

Notes 233

21 Alex Callinicos, Theories and Narratives: Reflections on the Philosophy ofHistory (Cambridge, Polity Press, 1995), pp. 95–96.

22 Paul Ricoeur, Time and Narrative (trans. K. McLaughlin and D. Pellauer,1, Chicago, University of Chicago Press, 1983–84), pp. 130–31.

23 Philippe Carrard, Poetics of the New History: French Historical Discoursefrom Braudel to Chartier (Baltimore, Johns Hopkins University Press,1992), pp. 74–82, esp. p. 75.

24 Robert F. Berkhofer, Beyond the Great Story: History as Text and Discourse(Cambridge, Massachusetts, Harvard University Press, 1995), p. 58.

25 Callinicos, Theories and Narratives, op. cit., p. 76.26 Lawrence Stone, ‘History and Post-Modernism’, Past and Present, No. 131,

1991, p. 191.27 Ibid., p. 192.28 Simon Schama, Landscape and Memory (London, HarperCollins, 1995),

p. 624.29 Marshal Sahlins, Islands of History (Chicago, University of Chicago

Press, 1985).30 Burke, New Perspectives on Historical Writing, op. cit., p. 240.31 Ibid., pp. 240–241.32 Ibid., p. 241.33 Claire Sanders in interview with Natalie Zemon Davis, ‘The Truth About

Fiction’, Times Higher Education Supplement, 10 November 1995, p. 21.34 Schama, Landscape and Memory, op. cit., p. 7.35 James A. Henretta, ‘Social History as Lived and Written’, American

Historical Review, Vol. 84, No. 5, December 1979, pp. 1318–1319.36 Ibid.37 I explore the idea of history as representing cultural memory in Alun Muns-

low, ‘Imagining the Nation: The Frontier Thesis and the Creating of Amer-ica’, in Philip John Davies (ed.), Representing and Imagining America(Keele, Keele University Press, 1996), pp. 15–23.

38 F.R. Ankersmit, ‘Historiography and Postmodernism’, History and Theory,Vol. 28, No. 2, 1989, p. 152.

39 R.G. Collingwood, The Idea of History (Oxford, Oxford University Press,revised edition, 1994), p. 434.

40 David Carroll, ‘Poetics, Theory, and the Defence of History’, Clio, Vol. 22,No. 3, 1993, pp. 273–289.

7 MICHEL FOUCAULT AND HISTORY

1 For a definitive listing of Foucault’s work see James Bernauer and ThomasKeenan, ‘The Works of Michel Foucault, 1954–1984’, in James Bernauerand David Rasmussen (eds), The Final Foucault (Cambridge, Mas-sachusetts, MIT Press, 1988). Among the most accessible commentaries onFoucault the historian are Hayden White, ‘Structuralism and Popular Cul-ture’, Journal of Popular Culture, Vol. 7, 1974, pp. 759–775 and ‘FoucaultDecoded: Notes From Underground’, History and Theory, Vol. 12, 1973,pp. 23–54, reprinted in Tropics of Discourse: Essays in Cultural Criticism(Baltimore, Johns Hopkins University Press, 1978). See also Hubert L.Dreyfus and Paul Rabinow, Michel Foucault: Beyond Structuralism and

234 Notes

Hermeneutics (Brighton, Harvester Press, second edition, 1983); MarkPoster, Foucault, Marxism and History (London, Polity Press, 1984); J.G.Merquior, Foucault (London, Fontana, 1985); Allan Megill, Prophets ofExtremity: Nietzsche, Heidegger, Foucault, Derrida (Berkeley, University ofCalifornia Press, 1985) and ‘The Reception of Foucault by Historians’,Journal of the History of Ideas, Vol. 48, 1987, pp. 117–141; Gary Gutting,The Cambridge Companion to Foucault (Cambridge, Cambridge UniversityPress, 1994); Lois McNay, Foucault, A Critical Introduction (New York,Continuum, 1994); Alan Sheridan, Michel Foucault, The Will to Truth(London, Routledge, reprinted 1994); Gerard Noiriel, ‘Foucault and His-tory: The Lessons of a Disillusion’, Journal of Modern History, Vol. 66,September 1994, pp. 547–568 and Michael S. Roth, The Ironist’s Cage:Memory, Trauma and the Construction of History (New York, ColumbiaUniversity Press, 1995), pp. 71–136.

2 Rudy Koshar, ‘Foucault and Social History: Comments on “CombinedUnderdevelopment” ’, American Historical Review, Vol. 98, No. 2, April1993, p. 358.

3 Michel Foucault, The Order of Things: An Archaeology of the HumanSciences (New York, Random House, 1973).

4 White, ‘Foucault Decoded’, in Tropics of Discourse, op. cit.5 Quoted in Noiriel, ‘Foucault and History’, op. cit., p. 551.6 Roth, The Ironist’s Cage, op. cit., pp. 72–78 and Clayton Roberts, The Logic

of Historical Explanation (University Park, University of PennsylvaniaPress, 1996), pp. 183–192.

7 Roth, The Ironist’s Cage, op. cit., p. 76.8 One of the best analyses of Foucault’s epistemology is to be found in Drey-

fus and Rabinow, Michel Foucault, op. cit., pp. 124–125. Michel Foucault,‘Nietzsche, Genealogy, History’, in Language, Counter Memory, Practice:Selected Essays and Interviews, ed. by Donald F. Bouchard, and trans. byDonald F. Bouchard and Sherry Simon (Ithaca, Cornell University Press,1977), pp. 139–164.

9 Ibid., p. 157.10 Ibid., p. 158.11 Michel Foucault, The Archaeology of Knowledge (New York, Harper &

Row, 1972).12 Michel Foucault, The Birth of the Clinic: An Archaeology of Medical Per-

ception (New York, Vintage Books, 1975) and Madness and Civilization: AHistory of Insanity in the Age of Reason (London, Tavistock, 1973).

13 Patrick Joyce, Democratic Subjects: The Self and the Social in NineteenthCentury England (Cambridge, Cambridge University Press, 1994), p. 9.

14 White, ‘Structuralism and Popular Culture’, op. cit., p. 771.15 Quoted in Lynn Hunt (ed.), The New Cultural History (Berkeley, University

of California Press, 1989), p. 7.16 Foucault, The Archaeology of Knowledge, op. cit., p. 191.17 Ibid.18 Ibid.19 White, ‘The Tropics of History’ and ‘Foucault Decoded’ in Tropics of Dis-

course, op. cit., pp. 254, 197.20 Ibid.21 Louis Althusser, Lenin and Philosophy and Other Essays (New York,

Notes 235

Monthly Review Press, 1971), p. 162. For a lengthier introduction see AlunMunslow, Discourse and Culture: The Creation of America, 1870–1920(London, Routledge, 1992), pp. 177–178.

22 Poster, Foucault, Marxism and History, op. cit., p. 71.

8 HAYDEN WHITE AND DECONSTRUCTIONIST HISTORY

1 John Passmore, ‘Explanation in Everyday Life, in Science, and in History’,History and Theory, Vol. 2, No. 2, 1962, pp. 122, 123, quoted by G. Roberts,‘Narrative History as a Way of Life’, Journal of Contemporary History, Vol.31, 1996, pp. 221–228. This issue also contains responses and replies to theMarwick–White dialogue.

2 F.R. Ankersmit, History and Tropology: The Rise and Fall of Metaphor(Berkeley, University of California Press, 1994), p. 3.

3 Keith Jenkins, On ‘What is History?’ (London, Routledge, 1995), pp. 134–179 and Keith Green and Jill LeBihan, Critical Theory and Practice: ACoursebook (London, Routledge, 1996), pp. 92–93, 100–101, 136–137. Seealso Raphael Samuel’s empiricist dismissal of White in Theatres of Memory(London, Verso, 1994), pp. 8, 41–42 and Roger Chartier, On the Edge ofthe Cli!: History, Language and Practice (Baltimore and London, JohnsHopkins University Press, 1997), pp. 28–38.

4 Frederick A. Olafson, ‘Hermeneutics: “Analytical” and “Dialectical” ’,History and Theory, Beiheft 25, 1986, pp. 28–42.

5 Hayden White, ‘The Politics of Historical Interpretation: Discipline andDe-Sublimation’, The Content of the Form: Narrative Discourse and Histor-ical Representation (Baltimore and London, Johns Hopkins UniversityPress, 1987), pp. 58–82.

6 Hayden White, ‘The Historical Text as Literary Artifact’, in Tropicsof Discourse: Essays in Cultural Criticism (Baltimore, Johns HopkinsUniversity Press, 1978), p. 85.

7 Hayden White, ‘Historicism, History and the Figurative Imagination’, inibid., pp. 101–120.

8 Hayden White, The Content of the Form: Narrative Discourse and HistoricalRepresentation (Baltimore, Johns Hopkins University Press, 1987), p. 209.

9 Robert Berkhofer, Beyond the Great Story: History as Text andDiscourse (Cambridge, Massachusetts, Harvard University Press, 1995),pp. 134–135.

10 Alun Munslow, Discourse and Culture: The Creation of America, 1870–1920(London, Routledge, 1992) and Dorothy Ross, ‘Grand Narratives inAmerican Historical Writing: From Romance to Uncertainty’, AmericanHistorical Review, Vol. 100, No. 3, June 1995, pp. 651–677.

11 Hayden White, ‘Interpretation in History’, ‘The Tropics of History: TheDeep Structure of the New Science’ and ‘Foucault Decoded: Notes FromUnderground’, in Tropics of Discourse, op. cit., pp. 51–80, 197–217, 230–260 and ‘Structuralism and Popular Culture’, Journal of Popular Culture,Vol. 7, 1974, pp. 759–775; Munslow, Discourse and Culture, op. cit., pp. 1–4.

12 Hayden White, ‘The Fictions of Factual Representation’, in Tropics ofDiscourse, op. cit., p. 134.

13 White, ‘The Historical Text as Literary Artifact’, op. cit., pp. 84–85.

236 Notes

14 Saul Friedlander, ‘Introduction’, in Saul Friedlander (ed.), Probing theLimits of Representation. Nazism and the ‘Final Solution’ (Cambridge andLondon, Harvard University Press, 1992), p. 3; Hayden White, ‘HistoricalEmplotment and the Problem of Truth’, in Probing the Limits of Represen-tation. Nazism and the ‘Final Solution’ (Cambridge and London, HarvardUniversity Press, 1992), pp. 37–53.

15 Friedlander, op. cit.16 S. Monk, The Sublime (Ann Arbor, University of Michigan Press, 1960).17 White, ‘Interpretation in History’, op. cit., p. 60.18 Ankersmit, History and Tropology, op. cit., p. 41. See also Ankersmit, Sub-

lime, Historical Experience, op. cit.19 Ibid., pp. 34–36.20 White, ‘Interpretation in History’, op. cit., pp. 55–56.21 Jenkins, On ‘What is History?’, op. cit., p. 85, quoting from White, The

Content of the Form, op. cit., p. 1.22 White, ‘Interpretation in History’, op. cit., p. 73.23 Clayton Roberts, The Logic of Historical Explanation (University Park,

University of Pennsylvania Press, 1996) is only one recent attempt toreinstate positivism and covering laws in historical explanation.

24 Philippe Carrard, Poetics of the New History: French Historical DiscourseFrom Braudel to Chartier (Baltimore, Johns Hopkins University Press,1992), p. 75.

25 Alun Munslow, ‘Imagining the Nation: The Frontier Thesis and theCreating of America’, in Philip J. Davies (ed.), Representing and ImaginingAmerica (Keele, Keele University Press, 1996), pp. 15–23.

26 Hayden White, ‘Response to Arthur Marwick’, Journal of ContemporaryHistory, Vol. 30, No. 2, April 1995, p. 240.

27 Ibid., p. 244.28 Hayden White, Metahistory: The Historical Imagination in the Nineteenth

Century (Baltimore, Johns Hopkins University Press, 1973), p. 30.29 White, ‘Historicism, History and the Figurative Imagination’, op. cit.,

pp. 101–120.30 The best introduction to the White model is found in Jenkins, On ‘What is

History?’, op. cit., pp. 146–173.31 White, Metahistory, op. cit., p. 29.32 Ibid., pp. 29–30.33 White, ‘Interpretation in History’, op. cit., p. 71.34 Ibid., p. 72.35 White, Metahistory, op. cit., p. 34.36 Ibid.37 Ibid., p. 30.38 Michael S. Roth, The Ironist’s Cage: Memory, Trauma and the Construction

of History (New York, Columbia University Press, 1995), p. 144.39 White, ‘Interpretation in History’, op. cit., p. 73.40 White, Metahistory, op. cit., pp. 7–11.41 Ibid., p. 11.42 Ibid., p. 22.43 Emmanuel Levinas, Otherwise than Being: or, Beyond Essence, trans.

Alphonso Lingis (The Hague, Martinus Nijho!, [1973] 1981); Frank R.Ankersmit, ‘In Praise of Subjectivity’, in David Carr, Thomas R. Flynn,

Notes 237

and Rudolf A. Makkreel (eds), The Ethics of History (Evanston,Northwestern University Press, 2004), pp. 3–27.

44 White, Metahistory, op. cit., p. 24.45 Ibid.46 Louis Mink, ‘History and Fiction as Modes of Comprehension’, New

Literary History, Vol. 1, 1970, pp. 541–558.47 Ibid., pp. 557–558.48 Ankersmit, History and Tropology, op. cit., p. 72.

9 CONCLUSION

1 Peter De Bolla, ‘Disfiguring History’, in Suzanne Gearhart (ed.), The OpenBoundary of History and Fiction: A Critical Approach to the FrenchEnlightenment (Princeton, Princeton University Press, 1984), pp. 57–64 andAlun Munslow, Discourse and Culture: The Creation of America, 1870–1920(London, Routledge, 1992).

2 This is a view explicitly argued by Elizabeth Deeds Ermath in Sequel toHistory: Postmodernism and the Crisis of Historical Time (Princeton,Princeton University Press, 1992).

3 George A. Reisch, ‘Chaos Theory and Narrative’, History and Theory, Vol.30, No. 1, 1991, pp. 1–20, esp. p. 1.

4 Cushing Strout, ‘Border Crossings: History, Fiction, and Dead Certainties’,History and Theory, Vol. 31, No. 2, 1992, pp. 153–162.

5 Natalie Zemon Davis, ‘On the Lame’, American Historical Review, Vol. 93,No. 3, 1988, pp. 572–575. See also the attack on Davis’s The Return ofMartin Guerre in the same issue, Robert Finlay, ‘The Refashioning of Mar-tin Guerre’, pp. 553–571, in which Finlay describes the book as failing toreach the acceptable standards of reconstructionist historical scholarship.

6 C. Behan McCullagh, The Logic of History: Putting Postmodernism inPerspective (London and New York, Routledge, 2004), p. 194.

7 F.R. Ankersmit, History and Tropology: The Rise and Fall of Metaphor(Berkeley, University of California Press, 1994), p. 44.

8 Elizabeth Tonkin, ‘History and the Myth of Realism’, in Raphael Samueland Paul Thompson (eds), The Myths We Live By (London, Routledge,1990), p. 27.

9 Joyce Appleby, Lynn Hunt and Margaret Jacob, Telling the Truth AboutHistory (New York, Norton, 1994), p. 279.

10 Hayden White, ‘The Fictions of Factual Representation’, in Tropics of Dis-course: Essays in Cultural Criticism (Baltimore, Johns Hopkins UniversityPress, 1978), p. 134.

11 Philippe Carrard, Poetics of the New History: French Historical Discoursefrom Braudel to Chartier (Baltimore, Johns Hopkins University Press,1992), p. 18.

12 Quoted in Richard T. Vann, ‘Louis Mink’s Linguistic Turn’, History andTheory, Vol. 26, No. 1, 1987, p. 12.

13 Hayden White, Metahistory: The Historical Imagination in the NineteenthCentury (Baltimore, Johns Hopkins University Press, 1973), p. 31.

14 Hayden White, ‘The Tropics of History: The Deep Structure of the “NewScience” ’, in Tropics of Discourse, op. cit., p. 208.

238 Notes

15 Hayden White, ‘Response to Arthur Marwick’, Journal of ContemporaryHistory, Vol. 30, No. 2, April 1995, p. 239.

16 John Milton Cooper, Jnr, Pivotal Decades: The United States, 1900–1920(New York, Norton, 1990), p. 158.

17 White, Tropics of Discourse, op. cit., p. 90.

Notes 239

Index

aesthetics/aesthetic turn 7, 19, 46, 51,58, 61, 75, 102, 103, 109, 110, 154,159, 165, 166, 171, 172, 177, 192,197, 200, 201, 214

Althusser, L. 147American Civil War 7American Historical Association 68,

93, 100Americanisation 168anarchist 37, 165, 171–2Anderson, P. 21Ankersmit, F.R. 17, 21, 32, 33, 57, 58,

62, 66, 72, 75, 80, 84, 108, 111, 119,122, 123, 124, 149, 151, 155, 159,160, 171, 173, 177, 178, 184, 190,192, 211, 213

Annales 9, 24, 25, 26, 52, 53, 96, 118,139, 186, 193

Anthropological, the 142–3anthropology 10, 24, 26, 29, 51, 54,

97, 120, 121, 185, 211Appleby, J. 20, 41, 42, 44, 50, 51, 58,

82, 83, 86, 87, 91, 97, 102, 106, 180,210; see also Hunt, L.; Jacob, M.

a priori 20, 29, 44, 52, 63, 64, 80, 90,92, 93, 94, 119, 145, 180, 186, 193,195, 196, 197

argument 37, 38, 165, 166, 169, 170,185, 188, 198

Attridge, D. 210, 221

Bacon, F. 44, 63, 178, 196Bann, S. 64, 210, 220, 226, 229Barthes, R. 29, 32, 62–7 passim, 75,

78, 85, 121, 155, 178, 194, 200, 203Beard, C. 25, 69, 88, 110, 184

Becker, C. 69, 88, 110, 184Bennington, G. 210Bentley, M. 208Berger, S. 208Berkhofer, R.F. 100, 118, 119, 154,

210Bernauer, J. 213Bernstein, R.J. 80Bevir, M. 84, 111–12, 114, 209Bloch, M. 25, 52Braudel, F. 25, 53, 54, 211Braverman, H. 54Breisach, E. 211Brown, C.G. 210Bruner, J. 109, 212Burckhardt, J. 166, 192, 211Burke, P. 49, 114, 121, 122, 211Burns, R.M. 209

Cabrera, M.A. 210Callinicos, A. 21, 44, 49, 50, 51, 60,

68, 104, 118, 119, 211Canary, R. 211Cannadine, D. 208Carr, D. 13, 72, 108, 125, 210, 211,

212Carr, E.H. 41, 48, 49, 50, 53, 68, 86,

89, 90, 91, 92, 97, 99, 105, 147, 181,184, 187, 208

Carrard, P. 53, 57, 91, 118, 122, 162,185

Carroll, D. 101, 102causality 7, 12, 34, 49, 58, 59, 62, 63,

66, 68, 74, 84, 89, 96, 100, 102, 103,118, 121, 125, 127, 129, 140, 141,148, 150, 193, 194, 196, 201

Chaos Theory 15Chartier, R. 21, 27, 53, 54, 114, 139,

210Chartists 173Chatman, S. 109, 212class 9, 11, 12, 26, 47, 49, 50, 51, 52,

54, 57, 59, 78, 112, 120, 140, 143,150, 157, 163, 169, 184, 193, 195,199

Classical, the 142cliometrics 96, 156, 193Cold War 15, 73, 83, 126, 173, 182,

183colligation 8, 60, 164, 170, 181, 193,

196, 198, 201Collingwood, R.G. 9, 12, 22, 23, 48,

56, 67, 68, 69, 90, 91, 92, 99, 100,101, 104, 105, 110, 112, 124, 125,147, 184, 187, 195, 198

Columbus 173comedy 13, 37, 102, 109, 153, 160,

165, 166, 168, 169comic see comedycomic strip 77Comte, A. 24, 202conservatism 165, 166, 171, 172constructionist history 4, 9, 10, 11,

20, 21, 24–7, 34, 39–60, 68, 69, 70,71, 76, 82, 83, 86, 93, 94, 95, 96,97, 98, 99, 103, 104, 105, 107–128,131, 137, 138, 139, 140, 143, 146,149, 151, 152, 154, 157, 159,160–4 passim, 168, 175, 178, 179,180, 181, 184, 185, 186, 193, 195,198, 201, 202, 203, 205, 206, 208,211

content and form 61, 80, 116contextualism/contextualisation 20,

26, 28, 45, 48, 57, 58, 62, 66, 74, 75,80, 88, 92

contextualist 165, 166correspondence theory 10, 14, 22, 28,

29, 35, 41, 42, 51, 56, 62, 83, 85, 86,92, 116, 133, 135, 139, 143, 148,152, 176, 188, 194, 202, 203, 204

covering law 49, 53, 67, 68, 69, 70, 83,94, 118, 151, 162, 164, 176, 184,186, 194, 202

Croce, B. 22, 23, 69, 184, 203cultural signature 146

Daddow, O. 213Danto, A. 6, 12, 57, 72, 102, 211, 216Darnton, R. 21, 25, 54, 120Davies, S. 208Davis, M. 26Davis, N.Z. 25, 120, 121–2, 176, 177Dean, M. 213Deane, P. 59death of the author 66, 88, 135, 194deconstructionism 31, 94, 95, 100,

103, 110, 118, 194, 200, 202deconstructionist history 10, 16, 18,

20, 21, 24, 26, 27, 28, 29, 33, 34,35–8, 44, 48, 49, 58, 61–106, 108,111, 113, 114, 116, 117, 118, 119,124, 127, 132, 135, 136, 139, 147,149–74, 175, 176, 177, 179, 180,182, 183, 184, 186, 189, 190, 193,194, 195, 199, 203, 204, 205, 208

deconstructive turn 72, 80, 86, 91,107, 119, 124, 145

deduction 5, 39, 41, 42, 44, 47, 48, 49,52, 53, 87, 88, 93, 124, 129, 142,146, 180, 184, 186, 193, 199, 201

Deleuze, G. 131Derrida, J. 3, 24, 27, 32, 62, 79, 85, 87,

97, 112, 113, 121, 177, 178, 194,195, 200, 203, 204, 210

Descartes 66, 196, 201diachronic 30, 141, 186di!érance (di!erence) 32, 195Dilthey, W. 198Downs, L.L. 211Dray, W.H. 71, 209Dreyfus, H.L. 213Dubois, B.C. 54

Easthope, A. 136Elias, A.J. 71Elias, N. 21, 96Elton, G. 20, 22, 23, 24, 25, 26, 42,

44–60, 65, 67, 68, 82, 83, 86, 87, 88,90, 91, 92, 93, 94, 105, 119, 148,180, 181, 182, 207

empathy 9, 91, 99, 110, 147, 184, 195,198

empiricism 4, 10, 11, 14, 19, 20, 21,22, 24, 25, 29, 31, 33, 38, 39–60, 61,62, 63, 67, 71, 74, 76, 83, 87, 89, 92,94, 101, 105, 108, 111–13, 124, 125,

242 Index

129, 133, 145, 150, 154, 155, 157,160, 167, 174, 177, 178, 184, 185,189, 192, 193, 194, 195, 196, 197,201, 202, 203, 204, 206, 208

emplotment 6, 8, 11, 13, 14, 20, 34, 35,36, 37, 38, 45, 54, 56, 57, 75, 77, 79,80, 86, 100, 102, 103, 105, 109, 115,116, 118, 119, 120, 126, 144, 145,146, 153, 154, 155, 157, 159, 161,164, 165, 166, 168, 169, 170, 181,185, 187, 188, 189, 193, 196, 197,198, 199, 201, 206

Engelstein, L. 214Enlightenment, the 2, 16, 52, 96, 146,

158, 196, 200, 201, 202, 209episteme 77, 129–48 passim, 151, 154,

155, 161, 168, 179, 186, 188, 189,196, 197, 201

epistemology 4, 5, 18, 20, 22, 36, 38,39–45, 60, 62–6, 76, 81, 82–8,107–22, 129, 130–5, 143, 151–5,162, 175, 176, 178, 197

ethical turn see ethicsethics 11, 116, 138, 159, 166, 170, 171,

185, 187, 197, 205, 210, 213, 215Evans-Pritchard, E.E. 96evidence 2–18, 20, 24–8, 32, 34–6,

39–52, 56, 57, 60–1, 64–70, 72–6,79–80, 82, 84, 87–96, 98, 99, 100,102, 104, 105, 106–10, 112–18, 119,122, 124–7, 131–41, 143, 145, 147,148, 149, 150, 152, 155–61, 163,164, 173, 175–85, 187, 188, 189,194, 195, 197, 198, 199, 202, 203,207

fact(s) 2, 3, 5, 7–9, 18, 36, 38, 41, 44,45, 47, 48, 49, 50, 53, 60, 64, 65, 67,69, 73, 75, 77, 80, 86, 87, 88, 89,90–5, 97, 104, 105, 109, 113, 115,119, 120, 124, 133, 135, 136, 143,155, 156–61, 172, 180, 181, 183,184, 187, 188, 194, 198, 202, 203,204

farce 101, 126, 160fascist history 14, 17Febvre, G. 52Feldner, H. 208feminist history 23, 54, 158fiction 11, 13, 27, 31, 34, 36, 41, 47,

60, 62, 64, 65, 69, 74, 75, 79, 80, 83,85, 97, 99, 104, 105, 110, 120, 121,122, 126, 133, 152, 153, 158, 161,173, 176, 180, 185, 194, 196, 202,212

Flynn, T.R. 210Foner, P. 26formist 37, 165, 169Foucault, M. 10, 14, 15, 16, 29, 31, 35,

36, 37, 42, 63, 65, 70, 75, 77, 78, 79,81, 85, 103, 107, 120, 121, 125, 128,129–48 passim, 151, 153, 154, 155,161, 162, 164, 166, 167, 174, 176,178, 179, 183, 184, 186, 187, 188,194, 196, 197, 200, 201, 203, 204,205, 210, 213, 214

free land 54, 68, 167–8French Revolution 16, 156Friedlander, S. 79, 157, 209Fulbrook, M. 208functionalism 96

Gadamer, H.G. 116, 203Gallie, W.B. 12, 13, 57, 59, 72, 122,

144, 149, 187, 211Gardiner, J. 208Gay, P. 19, 91, 192, 211Geertz, C. 54, 78, 96, 120, 122, 211Gellner, E. 54gender 11, 12, 26, 49, 50, 51, 54, 59,

108, 112, 113, 120, 184, 211genealogy 132, 133, 134, 138, 183Genette, G. 109, 212Genovese, E. 21, 54Giddens, A. 21, 54Ginsburg, C. 120, 176Goldstein, J. 213, 214Goldstein, L. 72, 88, 108, 111, 209Gordon, L. 86, 87, 180Graves, M.A.R. 43Great Chain of Being 142Green, A. 207Gutman, H. 26, 54Gutting, G. 214

Hall, C. 54Halley’s Comet 189Handlin, O. 20, 44Hardy, B. 172–3Harlan, D. 21, 72, 116, 117, 183, 209

Index 243

Harvey, D. 210Hegel, F. 16, 61, 66, 166, 200Heidegger, M. 198, 203Hempel, C. 53, 68, 194Henretta, J.A. 122, 123hermeneutics 81, 82, 86, 87, 88, 107,

136, 152, 155, 195, 198, 213Hesse, C. 10, 20, 208heteroglossia 114Hexter, J.H. 20, 23, 44, 55, 56, 57,

72Hill, C. 26Himmelfarb, G. 20, 44, 88, 95, 178,

208historical discourse 32–3, 126, 153,

170, 175historical explanation 1, 4, 5, 9, 11,

12, 14, 20, 32, 37, 39, 41, 44, 49, 51,56, 58, 62, 68, 69, 71, 72, 75, 78, 82,96, 103, 107, 108, 120, 146, 148,150, 151, 163, 164, 165, 169, 170,179, 185, 190, 194, 197, 199, 204,207, 210

historical interpretation 6, 8, 9, 31, 37,42, 43, 48, 51, 56, 61, 64, 67, 71, 75,82, 87, 92, 95, 100, 101, 110, 112,115, 123, 127, 132, 137, 158, 161,165, 173, 181, 185, 189, 198, 199

historical truth 1, 27, 34, 36, 41, 65,67, 72, 83, 86, 88, 97, 100, 105, 109,132, 133, 135, 178, 180, 183, 184,190, 203

historical understanding 4, 7, 12, 13,15, 16, 21, 26, 55, 57, 71, 80, 98, 99,102, 105, 111, 118, 119, 123, 147,152, 153, 166, 174, 175, 178, 184,187, 188, 193, 201, 211, 214

historicism 9, 164historiography 9, 18, 109, 156, 178,

208, 210, 212, 214History and Theory, 75Hobsbawm, E. 54, 104, 212Ho!er, P.C. 208Hollinger, D. 44, 102–3Holocaust 79, 104, 156, 157Hoskins, W.G. 54Hunt, L. 20, 41, 42, 44, 50, 51, 58, 82,

83, 85, 86, 87, 88, 91, 102, 106, 180,182, 207, 209; see also Appleby, J.;Jacob, M.

hypothesis testing 5, 118, 162, 185,187

idealism 94ideology 15, 22, 23, 31, 38, 46, 51, 61,

65, 67, 70, 74, 78, 79, 80, 81, 85, 89,92, 116, 119, 127, 133, 134, 139,166, 168, 171, 172, 174, 183, 194,198, 199

Iggers, G. 210, 214imperialism 28, 47impositionalism (of the historian) 23,

28, 42, 44, 48, 51, 63, 67, 89, 94, 98,99, 103, 112, 120, 123, 124, 127,128, 133, 146, 155, 161, 169, 171,174, 178, 179, 184, 185, 198, 199,202

induction 41, 44, 45, 69, 199Industrial Revolution 59, 73, 140inference 8, 39, 40, 41, 42, 44, 45, 46,

47, 49, 50, 52, 54, 56, 60, 63, 69, 90,91, 93, 108, 124, 130, 136, 139, 143,148, 150, 155, 162, 180, 182, 197,198, 199, 206

intentionality 5, 25, 25, 51, 57, 66, 84,94, 111, 114, 135, 150, 155, 158,171, 177, 178, 180, 182, 194, 201

intertextuality 34, 78, 80, 98, 114, 115,127, 135, 153, 173, 174, 183, 201

irony 13, 14, 19, 37, 58, 78, 85, 88,102, 103, 145, 146, 147, 153, 154,155, 163, 165, 166, 167, 168, 205,210

Jacob, M. 20, 41, 4244, 50, 51, 58, 82,83, 85, 86, 87, 88, 91, 102, 106, 180,182, 207; see also Appleby, J.; Hunt,L.

Jenkins, K. 13, 17, 21, 42, 62, 178,199, 200, 211, 214

Jordanova, L. 208Joyce, P. 21, 26, 136

Kansteiner, W. 214Kaplan, S.L. 209Keenan, T. 213Kellner, H. 62, 72, 108, 109, 178, 211,

214Kennedy, J.F. 125Kiernan, V. 26

244 Index

Kincaid, J.R. 113, 114Kloppenberg, J. 23, 82, 87, 88, 180knower and known 52, 66, 68, 86, 89,

135, 162Kolko, G. 54Koshar, R. 214Kozicki, H. 211Kristeva, J. 203Kuhn, T. 77, 83, 92

LaCapra, D. 21, 72, 73, 92, 178, 184,209, 212

Ladurie, E. Le Roy 25, 53, 54, 120,176

Lambert, P. 208language 6, 7, 10, 11, 12, 14, 15, 19,

21, 23, 24, 26–36, 40, 42, 44, 46, 51,54–6, 58, 62–5, 69–80, 83–6, 89, 94,97, 98, 100–5, 109, 110, 112, 114,115, 117, 119, 121, 122, 124, 126,127, 130–34, 136–47, 149–74passim, 178, 183, 185, 187, 188, 189,190, 192, 194, 195, 200, 201, 203,204, 207, 210, 212

langue 30Lemon, M.C. 5, 7, 12, 56, 57, 66, 73,

122, 125, 126, 187, 209, 211Levinas, E. 171, 197Levi-Strauss, C. 164Lewin, J.E. 212liberalism 165, 171, 196linguistic turn 10, 18, 27, 33, 93, 99,

108, 119, 121, 136, 192, 200, 202,208, 209, 121

Lloyd, C. 53, 211logical positivism 85logocentrism 65, 94, 121, 129, 139,

186, 194, 200Lorenz, C. 23, 209Louch, A.R. 72Lyotard, J.-F. 16, 17, 63, 85, 200, 202

McCullagh, C.B. 23, 40, 41, 53, 60, 89,91, 92, 101, 102, 108, 114, 177, 180,208

MacIntyre, A. 104McLennan, G. 214McMillan, J. F. 82MacRaild, D. 208Madison, J. 7, 155

madness 122, 128–9, 171Makkreel, R.A. 210, 217Mandelbaum, M. 72Mannheim, K. 171Marchitello, H. 217Marwick, A. 20, 23, 24, 25, 42, 44, 72,

86, 87, 88, 89, 90, 105, 180, 207, 214Marx, K. 24, 99Marxist history 9, 21, 25, 29, 49, 50,

51, 54, 68, 94, 104, 118, 119, 147,156, 158, 162, 184, 186, 193, 207,212

master narratives 16, 17, 49, 63, 171,200

mechanist(ic) 37, 57, 165, 169, 170Megill, A. 21, 122, 178, 190memory 47, 85, 112, 120, 123, 130,

176, 210, 213Merquior, J.G. 213meta-narrative 9, 17, 85, 200metaphor 11, 13, 14, 34, 37, 63, 67, 72,

77, 78, 80, 94, 114, 115, 120, 122,123, 124, 131, 136, 137, 141, 145,147, 153, 154, 156, 160, 163, 165,166, 167, 168, 177, 179, 187, 189,190, 192, 197, 205, 209, 212

metonymy 13, 14, 37, 145, 146, 147,153, 165, 166, 167, 205

microhistory 121micronarrative 121, 122mimesis/mimetic 55, 87, 123, 189, 192,

205, 214Mink, L. 3, 12, 13, 71, 122, 149, 172,

173, 176, 184, 188Modern, the see Anthropological, theModernisation School 21, 25M/modernism/modernist 1, 2, 6, 9,

10, 11, 12, 17, 18, 20, 22, 23, 25, 37,108, 125, 129, 132, 142, 143, 144,146, 148, 157, 171, 176, 191, 200,202

Montgomery, D. 54Morgan, S. 214multiculturalism 182Munslow, A. 42, 208, 211, 214myth 29, 36, 64, 65, 70, 78, 88, 104,

110, 130, 132, 133, 145, 146, 159,168, 173, 178, 190, 205

narrative 201

Index 245

narrative turn 161new cultural history 10, 93, 112, 121,

122, 209New Economic History 10, 118new empiricism 10, 31, 71, 201, 208new historicism 26, 29, 33, 34, 97, 201New History 24, 52, 57, 95, 118, 133,

162, 208, 211Nietzsche, F. 63, 69, 85, 129, 132, 133,

139, 203, 210Noiriel, G. 214Norman, A. P. 102, 212Novick, P. 20, 87, 91, 111, 209

objectivity 1–18 passim, 20, 22, 23, 27,28, 29, 39, 40, 47, 48, 50, 52, 55, 60,63–7, 70–2, 75, 79, 82–8, 91–2, 94,95, 101–5, 108, 110–11, 113, 118,120, 123, 125, 128, 129, 132, 133,135, 138, 139, 143, 146, 151, 152,157, 164, 167, 175, 177, 178, 179,180, 183, 185, 190, 194, 195, 200,201, 202, 204, 205, 209

Olafson, F.A. 72, 82, 89, 90, 152organicist 37, 165, 166, 169, 170other, the 14, 85, 101, 111, 171, 178,

302

Palmer, B. 207paradigm(s) 3, 16, 18, 19, 24, 34, 46,

83, 85, 91, 95, 107, 108, 110, 124,126, 127, 132, 134, 159, 178, 180,208

parole 30Parrington, V.L. 25Partner, N. 211, 216part–part relationships 170part–whole relationships 136, 155Pencak, W. 30Pocock, J.G.A. 96Popper, K. 83, 85, 92positivist(ism) 3, 5, 26, 39, 49, 52, 53,

62, 68, 69, 70, 80, 83, 85, 97, 98, 99,101, 102, 107, 121, 122, 129, 139,156, 158, 184, 185, 186, 196, 200,202, 204, 209

Poster, M. 148, 213postmodern(ism/ist) 1–18 passim, 19,

21, 22, 27, 28, 33, 37, 41, 58, 63, 66,76, 82, 85, 86, 89, 93, 97, 102, 103,

104, 110, 111, 112, 126, 130, 133,138, 141, 143, 144, 145, 146, 148,176, 177, 190, 196, 200, 202, 208,209, 210, 211, 212, 213, 214, 215

post-structuralism 10, 19, 24, 27, 29,30, 31, 32–3, 35, 36, 62, 63, 71, 93,120, 121, 132, 148, 158, 203, 204,210

psychohistory 56, 95

Quellenkritik 48, 88, 90, 118, 120, 159,181

Rabinow, P. 213race 11, 12, 49, 50, 51, 54, 112, 120,

157, 169, 170, 184, 190radicalism 165, 171Ranke, L. von 22, 95, 110, 159, 192,

211Rasmussen, D. 213Rayment-Pickard, H. 209realistic see realityreality 64, 125, 147, 155, 183, 186,

189, 200, 202, 203reconstructionism 3, 12, 20, 22, 23,

24, 25, 26, 27, 28, 32, 34, 35, 36, 38,39–60, 61–3, 66, 67, 68, 70, 74, 76,80, 82, 87, 88, 89, 91, 92, 93, 94, 97,98, 102, 103, 104, 105, 107–28, 132,133, 136, 141, 143, 149, 150, 152,153, 158, 159, 160, 162, 164, 175,177, 178, 179, 180, 181, 184, 185,186, 193, 194, 195, 203, 204, 207,208

referentiality 5, 13, 30, 39, 41, 44, 62,65, 69, 70, 98, 128, 132, 149, 151,160, 163, 180, 190, 200, 202, 203,205

relativism 21, 22, 24, 28, 29, 36, 42, 43,48, 51, 52, 69, 70, 78, 79, 80, 83–92,95, 100, 104, 106, 110, 116–21, 132,162, 178, 181, 182, 184, 202, 203,204, 208

Renaissance, the 35, 96, 137, 142, 146,154

representation 2, 4, 8, 10, 11, 14, 17,18, 19, 21, 26, 27, 28, 33–8, 43, 51,58, 61, 67, 69, 70, 71, 75–6, 78, 82,84, 85, 92, 104, 107, 109, 117–19,120–3, 126, 128, 132, 134–8, 144–6,

246 Index

148, 151, 152, 153, 156, 157, 161,164, 166, 167, 171, 176, 177, 179,183, 187, 189, 190, 192, 193, 197,199, 200, 201, 202, 204, 209, 210,211, 212, 214

resemblance 13, 134, 137, 142, 204Rethinking History: The Journal of

Theory and Practice 76, 199, 210,211, 212, 213, 214, 215

rhetoric 11, 19, 34, 62, 64, 65, 75,77–9, 91, 101, 116, 124, 128, 137,139, 146, 150, 152, 154, 156, 157,158, 160–4, 166, 168, 170, 174, 176,177, 179, 183, 184, 185, 186, 192

Ricoeur, P. 13, 73, 75, 79, 109, 118,125, 149, 210

Roberts, C. 210Roberts, D.D. 210Roberts, D.M. 20, 210Robinson, J.H. 25, 52Roediger, D. 26romance 153, 160, 165, 196romantic see romanceRoosevelt, T. 189Rorty, R. 51, 87, 210Rosenstone, R.A. 211Ross, D. 213Roth, M. 132, 210Roth, P.A. 214Rowbotham, S. 54Royle, E. 43Rudé, G. 21, 26Ruiz, V. L. 59Rüsen, J. 62, 108, 178

Saari, H. 213Sahlins, M. 21, 120, 121satire 109, 160, 165, 166, 196satiric see satireSaussure de, F. 30, 77, 86, 97, 204, 205Saville, J. 26Schama, S. 120, 122, 176Schleirmacher, F. 198Schofield, P. 208scissors and paste 104Scott, J.W. 21, 107, 108Second World War 83semiotics 30, 95Sheridan, A. 214sign/signified/signifier 21, 30–2, 35,

54, 64–6, 69, 71, 73, 74, 76, 86, 89,90, 113, 119, 120, 127, 129, 130,131, 137–9, 151, 153, 159, 163, 166,167, 168, 170, 178, 182, 185, 186,189, 194, 195, 198, 203, 204

similarity 14, 137, 139, 167, 197Skinner, Q. 44, 96social construction of reality 115, 131social theory 4, 9, 20, 23, 24, 25, 26,

49, 51, 52, 54, 55, 56, 67, 68, 74, 82,88, 100, 103, 105, 107, 118, 120,121, 127, 139, 151, 161, 162, 163,164, 175, 186, 208, 211

socialism 17sociology 29, 51, 185, 202, 213Southgate, B. 210Sparks, C.D. 113, 114Spiegel, G.M. 20, 98, 114, 115, 116,

117, 208, 212Stanford, M. 23, 47, 47, 58, 59, 86, 87,

88, 104, 209Stone, L. 20, 24, 26, 34, 56, 57, 72, 95,

96, 97, 98, 119, 120, 185, 212structuralism 19, 24, 29–32, 56, 62, 73,

96, 120, 130, 141, 202, 203, 204, 213Stueck, W.W. 207style 12, 13, 19, 32, 34, 56, 62, 67, 102,

124, 125, 130, 151, 192, 202, 211subject/subjectivity 10, 15, 27, 50, 66,

67, 79, 85, 86, 95, 97, 102, 105, 123,139, 142, 147, 150, 151, 152, 177,189, 200

sublime nature of the past 6, 158, 162,178, 192, 211

synchronic 30, 141, 186synecdoche 13, 14, 37, 145, 146, 147,

153, 154, 165, 166, 167, 168, 189,205

Taylor, A. 208Taylor, A.J.P. 99, 100teleology 11, 55, 104, 129, 178tertium quid 76theories of history 7, 8, 48–55, 68–70,

93–9, 118–21, 139–43, 161–5, 180,184–7

Third World 21, 85Thompson, E.P. 21, 25, 26, 78Thompson, W. 78Tilly, C. 54

Index 247

Toews, J.E. 212Tolstoy, L. 112Tonkin, E. 179Topolski, J. 207Tosh, J. 20, 45, 82, 89, 92, 98, 99, 185,

207Toson, S. 121tragedy 13, 37, 101, 109, 126, 153,

165, 166, 168, 169, 170, 196tragic see tragedytranscendental signified 66, 69, 139Trebilcock, C. 59Trevor-Roper, Hugh 20, 44trompe l’œil 64, 87, 191trope 13, 14, 37, 38, 65, 77, 78, 86,

118, 119, 130, 134, 136, 142, 144,145, 146, 147, 151, 153, 154, 155,160, 161, 165, 166, 167, 172, 174,178, 179, 185, 188, 196, 205

truth-e!ect 44, 65, 74, 87, 102, 124,184

Turner, F.J. 24, 53, 68, 69, 93, 163,167, 168, 190, 191

Vann, R.T. 100, 187

verification 51, 60, 164, 181, 197Veyne, P. 13Vico, G. 140, 146–7, 167, 186–7

Walker, G. 211Walsh, W.H. 209Warsaw Uprising 101Weinstein, J. 54White, H. 11–15, 21, 24, 29, 33, 35–7,

42, 62, 65, 67, 70, 72, 73, 75, 77, 78,79, 80, 81, 86, 91, 100, 101, 102–4,108–9, 115, 119, 122, 123, 125, 126,128, 130, 132, 137, 138, 141, 144–74176, 178, 179, 183, 184, 185, 186,187, 188, 189, 190, 205, 209, 210,213, 214

whole–part relationships 146, 205Williams, R. 29, 35Winn, J.D. 83, 87, 105Wood, G.S. 20

Young, R. 210

Zagorin, P. 109Zeldin, T. 83

248 Index